nextmv 0.18.0__py3-none-any.whl → 1.0.0.dev2__py3-none-any.whl

nextmv/cloud/application/_batch_scenario.py

@@ -0,0 +1,860 @@
+"""
+Application mixin for managing batch experiments.
+"""
+
+from typing import TYPE_CHECKING, Any
+
+from nextmv.cloud.batch_experiment import (
+    BatchExperiment,
+    BatchExperimentInformation,
+    BatchExperimentMetadata,
+    BatchExperimentRun,
+    ExperimentStatus,
+    to_runs,
+)
+from nextmv.cloud.input_set import InputSet, ManagedInput
+from nextmv.cloud.scenario import Scenario, ScenarioInputType, _option_sets, _scenarios_by_id
+from nextmv.polling import DEFAULT_POLLING_OPTIONS, PollingOptions, poll
+from nextmv.run import Run
+from nextmv.safe import safe_id, safe_name_and_id
+
+if TYPE_CHECKING:
+    from . import Application
+
+
+class ApplicationBatchMixin:
+    """
+    Mixin class for managing batch experiments within an application.
+    """
+
+    def batch_experiment(self: "Application", batch_id: str) -> BatchExperiment:
+        """
+        Get a batch experiment. This method also returns the runs of the batch
+        experiment under the `.runs` attribute.
+
+        Parameters
+        ----------
+        batch_id : str
+            ID of the batch experiment.
+
+        Returns
+        -------
+        BatchExperiment
+            The requested batch experiment details.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> batch_exp = app.batch_experiment("batch-123")
+        >>> print(batch_exp.name)
+        'My Batch Experiment'
+        """
+
+        response = self.client.request(
+            method="GET",
+            endpoint=f"{self.experiments_endpoint}/batch/{batch_id}",
+        )
+
+        exp = BatchExperiment.from_dict(response.json())
+
+        runs_response = self.client.request(
+            method="GET",
+            endpoint=f"{self.experiments_endpoint}/batch/{batch_id}/runs",
+        )
+
+        runs = [Run.from_dict(run) for run in runs_response.json().get("runs", [])]
+        exp.runs = runs
+
+        return exp
+
+    def batch_experiment_metadata(self: "Application", batch_id: str) -> BatchExperimentMetadata:
+        """
+        Get metadata for a batch experiment.
+
+        Parameters
+        ----------
+        batch_id : str
+            ID of the batch experiment.
+
+        Returns
+        -------
+        BatchExperimentMetadata
+            The requested batch experiment metadata.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> metadata = app.batch_experiment_metadata("batch-123")
+        >>> print(metadata.name)
+        'My Batch Experiment'
+        """
+
+        response = self.client.request(
+            method="GET",
+            endpoint=f"{self.experiments_endpoint}/batch/{batch_id}/metadata",
+        )
+
+        return BatchExperimentMetadata.from_dict(response.json())
+
+    def batch_experiment_with_polling(
+        self: "Application",
+        batch_id: str,
+        polling_options: PollingOptions = DEFAULT_POLLING_OPTIONS,
+    ) -> BatchExperiment:
+        """
+        Get a batch experiment with polling.
+
+        Retrieves the result of an experiment. This method polls for the result
+        until the experiment finishes executing or the polling strategy is
+        exhausted.
+
+        Parameters
+        ----------
+        batch_id : str
+            ID of the batch experiment.
+
+        Returns
+        -------
+        BatchExperiment
+            The requested batch experiment details.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> batch_exp = app.batch_experiment_with_polling("batch-123")
+        >>> print(batch_exp.name)
+        'My Batch Experiment'
+        """
+
+        def polling_func() -> tuple[Any, bool]:
+            batch_metadata = self.batch_experiment_metadata(batch_id=batch_id)
+            if batch_metadata.status in {
+                ExperimentStatus.COMPLETED,
+                ExperimentStatus.FAILED,
+                ExperimentStatus.DRAFT,
+                ExperimentStatus.CANCELED,
+                ExperimentStatus.DELETE_FAILED,
+            }:
+                return batch_metadata, True
+
+            return None, False
+
+        batch_information = poll(polling_options=polling_options, polling_func=polling_func)
+
+        return self.batch_experiment(batch_id=batch_information.id)
+
+    def delete_batch_experiment(self: "Application", batch_id: str) -> None:
+        """
+        Delete a batch experiment.
+
+        Deletes a batch experiment along with all the associated information,
+        such as its runs.
+
+        Parameters
+        ----------
+        batch_id : str
+            ID of the batch experiment to delete.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> app.delete_batch_experiment("batch-123")
+        """
+
+        _ = self.client.request(
+            method="DELETE",
+            endpoint=f"{self.experiments_endpoint}/batch/{batch_id}",
+        )
+
+    def delete_scenario_test(self: "Application", scenario_test_id: str) -> None:
+        """
+        Delete a scenario test.
+
+        Deletes a scenario test. Scenario tests are based on the batch
+        experiments API, so this function summons `delete_batch_experiment`.
+
+        Parameters
+        ----------
+        scenario_test_id : str
+            ID of the scenario test to delete.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> app.delete_scenario_test("scenario-123")
+        """
+
+        self.delete_batch_experiment(batch_id=scenario_test_id)
+
+    def list_batch_experiments(self: "Application") -> list[BatchExperimentMetadata]:
+        """
+        List all batch experiments.
+
+        Returns
+        -------
+        list[BatchExperimentMetadata]
+            List of batch experiments.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        """
+
+        response = self.client.request(
+            method="GET",
+            endpoint=f"{self.experiments_endpoint}/batch",
+            query_params={"type": "batch"},
+        )
+
+        return [BatchExperimentMetadata.from_dict(batch_experiment) for batch_experiment in response.json()]
+
+    def list_scenario_tests(self: "Application") -> list[BatchExperimentMetadata]:
+        """
+        List all batch scenario tests. Scenario tests are based on the batch
+        experiments API, so this function returns the same information as
+        `list_batch_experiments`, albeit using a different query parameter.
+
+        Returns
+        -------
+        list[BatchExperimentMetadata]
+            List of scenario tests.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        """
+
+        response = self.client.request(
+            method="GET",
+            endpoint=f"{self.experiments_endpoint}/batch",
+            query_params={"type": "scenario"},
+        )
+
+        return [BatchExperimentMetadata.from_dict(batch_experiment) for batch_experiment in response.json()]
+
+    def new_batch_experiment(  # noqa: C901
+        self: "Application",
+        name: str | None = None,
+        input_set_id: str | None = None,
+        instance_ids: list[str] | None = None,
+        description: str | None = None,
+        id: str | None = None,
+        option_sets: dict[str, dict[str, str]] | None = None,
+        runs: list[BatchExperimentRun | dict[str, Any]] | None = None,
+        type: str | None = "batch",
+    ) -> str:
+        """
+        Create a new batch experiment.
+
+        Parameters
+        ----------
+        name: Optional[str]
+            Name of the batch experiment. If not provided, the ID will be used as the name.
+        input_set_id: str | None
+            ID of the input set to use for the batch experiment.
+        instance_ids: list[str]
+            This argument is deprecated, use `runs` instead.
+            List of instance IDs to use for the batch experiment.
+        description: Optional[str]
+            Optional description of the batch experiment.
+        id: Optional[str]
+            ID of the batch experiment. Will be generated if not provided.
+        option_sets: Optional[dict[str, dict[str, str]]]
+            Option sets to use for the batch experiment. This is a dictionary
+            where the keys are option set IDs and the values are dictionaries
+            with the actual options.
+        runs: Optional[list[BatchExperimentRun]]
+            List of runs to use for the batch experiment.
+        type: Optional[str]
+            Type of the batch experiment. This is used to determine the
+            experiment type. The default value is "batch". If you want to
+            create a scenario test, set this to "scenario".
+
+        Returns
+        -------
+        str
+            ID of the batch experiment.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        """
+
+        # Generate ID if not provided
+        if id is None:
+            id = safe_id("batch")
+
+        # Use ID as name if name not provided
+        if name is None:
+            name = id
+
+        payload = {
+            "id": id,
+            "name": name,
+        }
+        if input_set_id is not None:
+            payload["input_set_id"] = input_set_id
+        if instance_ids is not None:
+            input_set = self.input_set(input_set_id)
+            runs = to_runs(instance_ids, input_set)
+            payload_runs = [run.to_dict() for run in runs]
+            payload["runs"] = payload_runs
+        if description is not None:
+            payload["description"] = description
+        if option_sets is not None:
+            payload["option_sets"] = option_sets
+        if runs is not None:
+            payload_runs = [{}] * len(runs)
+            for i, run in enumerate(runs):
+                payload_runs[i] = run.to_dict() if isinstance(run, BatchExperimentRun) else run
+            payload["runs"] = payload_runs
+        if type is not None:
+            payload["type"] = type
+
+        response = self.client.request(
+            method="POST",
+            endpoint=f"{self.experiments_endpoint}/batch",
+            payload=payload,
+        )
+
+        return response.json()["id"]
+
+    def new_batch_experiment_with_result(
+        self: "Application",
+        name: str | None = None,
+        input_set_id: str | None = None,
+        instance_ids: list[str] | None = None,
+        description: str | None = None,
+        id: str | None = None,
+        option_sets: dict[str, dict[str, str]] | None = None,
+        runs: list[BatchExperimentRun | dict[str, Any]] | None = None,
+        type: str | None = "batch",
+        polling_options: PollingOptions = DEFAULT_POLLING_OPTIONS,
+    ) -> BatchExperiment:
+        """
+        Convenience method to create a new batch experiment and poll for the
+        result.
+
+        This method combines the `new_batch_experiment` and
+        `batch_experiment_with_polling` methods, applying polling logic to
+        check when the experiment succeeded.
+
+        Parameters
+        ----------
+        name: Optional[str]
+            Name of the batch experiment. If not provided, the ID will be used as the name.
+        input_set_id: str
+            ID of the input set to use for the batch experiment.
+        instance_ids: list[str]
+            List of instance IDs to use for the batch experiment. This argument
+            is deprecated, use `runs` instead.
+        description: Optional[str]
+            Optional description of the batch experiment.
+        id: Optional[str]
+            ID of the batch experiment. Will be generated if not provided.
+        option_sets: Optional[dict[str, dict[str, str]]]
+            Option sets to use for the batch experiment. This is a dictionary
+            where the keys are option set IDs and the values are dictionaries
+            with the actual options.
+        runs: Optional[list[BatchExperimentRun]]
+            List of runs to use for the batch experiment.
+        type: Optional[str]
+            Type of the batch experiment. This is used to determine the
+            experiment type. The default value is "batch". If you want to
+            create a scenario test, set this to "scenario".
+        polling_options : PollingOptions, default=DEFAULT_POLLING_OPTIONS
+            Options to use when polling for the batch experiment result.
+
+        Returns
+        -------
+        BatchExperiment
+            The completed batch experiment with results.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        """
+
+        batch_id = self.new_batch_experiment(
+            name=name,
+            input_set_id=input_set_id,
+            instance_ids=instance_ids,
+            description=description,
+            id=id,
+            option_sets=option_sets,
+            runs=runs,
+            type=type,
+        )
+
+        return self.batch_experiment_with_polling(batch_id=batch_id, polling_options=polling_options)
+
+    def new_scenario_test(
+        self: "Application",
+        scenarios: list[Scenario],
+        id: str | None = None,
+        name: str | None = None,
+        description: str | None = None,
+        repetitions: int | None = 0,
+    ) -> str:
+        """
+        Create a new scenario test. The test is based on `scenarios` and you
+        may specify `repetitions` to run the test multiple times. 0 repetitions
+        means that the tests will be executed once. 1 repetition means that the
+        test will be repeated once, i.e.: it will be executed twice. 2
+        repetitions equals 3 executions, so on, and so forth.
+
+        For each scenario, consider the `scenario_input` and `configuration`.
+        The `scenario_input.scenario_input_type` allows you to specify the data
+        that will be used for that scenario.
+
+        - `ScenarioInputType.INPUT_SET`: the data should be taken from an
+          existing input set.
+        - `ScenarioInputType.INPUT`: the data should be taken from a list of
+          existing inputs. When using this type, an input set will be created
+          from this set of managed inputs.
+        - `ScenarioInputType.New`: a new set of data will be uploaded as a set
+          of managed inputs. A new input set will be created from this set of
+          managed inputs.
+
+        On the other hand, the `configuration` allows you to specify multiple
+        option variations for the scenario. Please see the
+        `ScenarioConfiguration` class for more information.
+
+        The scenario tests uses the batch experiments API under the hood.
+
+        Parameters
+        ----------
+        scenarios: list[Scenario]
+            List of scenarios to use for the scenario test. At least one
+            scenario should be provided.
+        id: Optional[str]
+            ID of the scenario test. Will be generated if not provided.
+        name: Optional[str]
+            Name of the scenario test. If not provided, the ID will be used as the name.
+        description: Optional[str]
+            Optional description of the scenario test.
+        repetitions: Optional[int]
+            Number of repetitions to use for the scenario test. 0
+            repetitions means that the tests will be executed once. 1
+            repetition means that the test will be repeated once, i.e.: it
+            will be executed twice. 2 repetitions equals 3 executions, so on,
+            and so forth.
+
+        Returns
+        -------
+        str
+            ID of the scenario test.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        ValueError
+            If no scenarios are provided.
+        """
+
+        if len(scenarios) < 1:
+            raise ValueError("At least one scenario must be provided")
+
+        # Generate ID if not provided
+        if id is None:
+            id = safe_id("scenario")
+
+        # Use ID as name if name not provided
+        if name is None:
+            name = id
+
+        scenarios_by_id = _scenarios_by_id(scenarios)
+
+        # Save all the information needed by scenario.
+        input_sets = {}
+        instances = {}
+        for scenario_id, scenario in scenarios_by_id.items():
+            instance = self.instance(instance_id=scenario.instance_id)
+
+            # Each scenario is associated to an input set, so we must either
+            # get it or create it.
+            input_set = self.__input_set_for_scenario(scenario, scenario_id)
+
+            instances[scenario_id] = instance
+            input_sets[scenario_id] = input_set
+
+        # Calculate the combinations of all the option sets across scenarios.
+        opt_sets_by_scenario = _option_sets(scenarios)
+
+        # The scenario tests results in multiple individual runs.
+        runs = []
+        run_counter = 0
+        opt_sets = {}
+        for scenario_id, scenario_opt_sets in opt_sets_by_scenario.items():
+            opt_sets = {**opt_sets, **scenario_opt_sets}
+            input_set = input_sets[scenario_id]
+            scenario = scenarios_by_id[scenario_id]
+
+            for set_key in scenario_opt_sets.keys():
+                inputs = input_set.input_ids if len(input_set.input_ids) > 0 else input_set.inputs
+                for input in inputs:
+                    input_id = input.id if isinstance(input, ManagedInput) else input
+                    for repetition in range(repetitions + 1):
+                        run_counter += 1
+                        run = BatchExperimentRun(
+                            input_id=input_id,
+                            input_set_id=input_set.id,
+                            instance_id=scenario.instance_id,
+                            option_set=set_key,
+                            scenario_id=scenario_id,
+                            repetition=repetition,
+                            run_number=f"{run_counter}",
+                        )
+                        runs.append(run)
+
+        return self.new_batch_experiment(
+            id=id,
+            name=name,
+            description=description,
+            type="scenario",
+            option_sets=opt_sets,
+            runs=runs,
+        )
+
+    def new_scenario_test_with_result(
+        self: "Application",
+        scenarios: list[Scenario],
+        id: str | None = None,
+        name: str | None = None,
+        description: str | None = None,
+        repetitions: int | None = 0,
+        polling_options: PollingOptions = DEFAULT_POLLING_OPTIONS,
+    ) -> BatchExperiment:
+        """
+        Convenience method to create a new scenario test and poll for the
+        result.
+
+        This method combines the `new_scenario_test` and
+        `scenario_test_with_polling` methods, applying polling logic to
+        check when the test succeeded.
+
+        The scenario tests uses the batch experiments API under the hood.
+
+        Parameters
+        ----------
+        scenarios: list[Scenario]
+            List of scenarios to use for the scenario test. At least one
+            scenario should be provided.
+        id: Optional[str]
+            ID of the scenario test. Will be generated if not provided.
+        name: Optional[str]
+            Name of the scenario test. If not provided, the ID will be used as the name.
+        description: Optional[str]
+            Optional description of the scenario test.
+        repetitions: Optional[int]
+            Number of repetitions to use for the scenario test. 0
+            repetitions means that the tests will be executed once. 1
+            repetition means that the test will be repeated once, i.e.: it
+            will be executed twice. 2 repetitions equals 3 executions, so on,
+            and so forth.
+        polling_options : PollingOptions, default=DEFAULT_POLLING_OPTIONS
+            Options to use when polling for the scenario test result.
+
+        Returns
+        -------
+        BatchExperiment
+            The completed scenario test as a BatchExperiment.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        ValueError
+            If no scenarios are provided.
+        """
+
+        test_id = self.new_scenario_test(
+            scenarios=scenarios,
+            id=id,
+            name=name,
+            description=description,
+            repetitions=repetitions,
+        )
+
+        return self.scenario_test_with_polling(
+            scenario_test_id=test_id,
+            polling_options=polling_options,
+        )
+
+    def scenario_test(self: "Application", scenario_test_id: str) -> BatchExperiment:
+        """
+        Get a scenario test.
+
+        Retrieves a scenario test by ID. Scenario tests are based on batch
+        experiments, so this function returns the corresponding batch
+        experiment associated with the scenario test.
+
+        Parameters
+        ----------
+        scenario_test_id : str
+            ID of the scenario test to retrieve.
+
+        Returns
+        -------
+        BatchExperiment
+            The scenario test details as a batch experiment.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> test = app.scenario_test("scenario-123")
+        >>> print(test.name)
+        'My Scenario Test'
+        >>> print(test.type)
+        'scenario'
+        """
+
+        return self.batch_experiment(batch_id=scenario_test_id)
+
+    def scenario_test_metadata(self: "Application", scenario_test_id: str) -> BatchExperimentMetadata:
+        """
+        Get the metadata for a scenario test, given its ID.
+
+        Scenario tests are based on batch experiments, so this function returns
+        the corresponding batch experiment metadata associated with the
+        scenario test.
+
+        Parameters
+        ----------
+        scenario_test_id : str
+            ID of the scenario test to retrieve.
+
+        Returns
+        -------
+        BatchExperimentMetadata
+            The scenario test metadata as a batch experiment.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> metadata = app.scenario_test_metadata("scenario-123")
+        >>> print(metadata.name)
+        'My Scenario Test'
+        >>> print(metadata.type)
+        'scenario'
+        """
+
+        return self.batch_experiment_metadata(batch_id=scenario_test_id)
+
+    def scenario_test_with_polling(
+        self: "Application",
+        scenario_test_id: str,
+        polling_options: PollingOptions = DEFAULT_POLLING_OPTIONS,
+    ) -> BatchExperiment:
+        """
+        Get a scenario test with polling.
+
+        Retrieves the result of a scenario test. This method polls for the
+        result until the test finishes executing or the polling strategy is
+        exhausted.
+
+        The scenario tests uses the batch experiments API under the hood.
+
+        Parameters
+        ----------
+        scenario_test_id : str
+            ID of the scenario test to retrieve.
+        polling_options : PollingOptions, default=DEFAULT_POLLING_OPTIONS
+            Options to use when polling for the scenario test result.
+
+        Returns
+        -------
+        BatchExperiment
+            The scenario test details as a batch experiment.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> test = app.scenario_test_with_polling("scenario-123")
+        >>> print(test.name)
+        'My Scenario Test'
+        >>> print(test.type)
+        'scenario'
+        """
+
+        return self.batch_experiment_with_polling(batch_id=scenario_test_id, polling_options=polling_options)
+
+    def update_batch_experiment(
+        self: "Application",
+        batch_experiment_id: str,
+        name: str | None = None,
+        description: str | None = None,
+    ) -> BatchExperimentInformation:
+        """
+        Update a batch experiment.
+
+        Parameters
+        ----------
+        batch_experiment_id : str
+            ID of the batch experiment to update.
+        name : Optional[str], default=None
+            Optional name of the batch experiment.
+        description : Optional[str], default=None
+            Optional description of the batch experiment.
+
+        Returns
+        -------
+        BatchExperimentInformation
+            The information with the updated batch experiment.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+        """
+
+        payload = {}
+
+        if name is not None:
+            payload["name"] = name
+        if description is not None:
+            payload["description"] = description
+
+        response = self.client.request(
+            method="PATCH",
+            endpoint=f"{self.experiments_endpoint}/batch/{batch_experiment_id}",
+            payload=payload,
+        )
+
+        return BatchExperimentInformation.from_dict(response.json())
+
+    def update_scenario_test(
+        self: "Application",
+        scenario_test_id: str,
+        name: str | None = None,
+        description: str | None = None,
+    ) -> BatchExperimentInformation:
+        """
+        Update a scenario test.
+
+        Updates a scenario test with new name and description. Scenario tests
+        use the batch experiments API, so this method calls the
+        `update_batch_experiment` method, and thus the return type is the same.
+
+        Parameters
+        ----------
+        scenario_test_id : str
+            ID of the scenario test to update.
+        name : Optional[str], default=None
+            Optional new name for the scenario test.
+        description : Optional[str], default=None
+            Optional new description for the scenario test.
+
+        Returns
+        -------
+        BatchExperimentInformation
+            The information about the updated scenario test.
+
+        Raises
+        ------
+        requests.HTTPError
+            If the response status code is not 2xx.
+
+        Examples
+        --------
+        >>> info = app.update_scenario_test(
+        ...     scenario_test_id="scenario-123",
+        ...     name="Updated Test Name",
+        ...     description="Updated description for this test"
+        ... )
+        >>> print(info.name)
+        'Updated Test Name'
+        """
+
+        return self.update_batch_experiment(
+            batch_experiment_id=scenario_test_id,
+            name=name,
+            description=description,
+        )
+
+    def __input_set_for_scenario(self: "Application", scenario: Scenario, scenario_id: str) -> InputSet:
+        # If working with an input set, there is no need to create one.
+        if scenario.scenario_input.scenario_input_type == ScenarioInputType.INPUT_SET:
+            input_set = self.input_set(input_set_id=scenario.scenario_input.scenario_input_data)
+            return input_set
+
+        # If working with a list of managed inputs, we need to create an
+        # input set.
+        if scenario.scenario_input.scenario_input_type == ScenarioInputType.INPUT:
+            name, id = safe_name_and_id(prefix="inpset", entity_id=scenario_id)
+            input_set = self.new_input_set(
+                id=id,
+                name=name,
+                description=f"Automatically created from scenario test: {id}",
+                maximum_runs=20,
+                inputs=[
+                    ManagedInput.from_dict(data={"id": input_id})
+                    for input_id in scenario.scenario_input.scenario_input_data
+                ],
+            )
+            return input_set
+
+        # If working with new data, we need to create managed inputs, and then,
+        # an input set.
+        if scenario.scenario_input.scenario_input_type == ScenarioInputType.NEW:
+            managed_inputs = []
+            for data in scenario.scenario_input.scenario_input_data:
+                upload_url = self.upload_url()
+                self.upload_data(data=data, upload_url=upload_url)
+                name, id = safe_name_and_id(prefix="man-input", entity_id=scenario_id)
+                managed_input = self.new_managed_input(
+                    id=id,
+                    name=name,
+                    description=f"Automatically created from scenario test: {id}",
+                    upload_id=upload_url.upload_id,
+                )
+                managed_inputs.append(managed_input)
+
+            name, id = safe_name_and_id(prefix="inpset", entity_id=scenario_id)
+            input_set = self.new_input_set(
+                id=id,
+                name=name,
+                description=f"Automatically created from scenario test: {id}",
+                maximum_runs=20,
+                inputs=managed_inputs,
+            )
+            return input_set
+
+        raise ValueError(f"Unknown scenario input type: {scenario.scenario_input.scenario_input_type}")