nominal 1.105.0__tar.gz → 1.107.0__tar.gz

{nominal-1.105.0 → nominal-1.107.0}/CHANGELOG.md

@@ -1,5 +1,24 @@
 # Changelog
 
+## [1.107.0](https://github.com/nominal-io/nominal-client/compare/v1.106.0...v1.107.0) (2026-01-22)
+
+
+### Features
+
+* make migration preserve uuid and dataset include options configurable ([#584](https://github.com/nominal-io/nominal-client/issues/584)) ([95ae64d](https://github.com/nominal-io/nominal-client/commit/95ae64de267ad957d15641378b6b418e8c234faa))
+
+## [1.106.0](https://github.com/nominal-io/nominal-client/compare/v1.105.0...v1.106.0) (2026-01-21)
+
+
+### Features
+
+* create dataset with uuid endpoint for mgiration ([#580](https://github.com/nominal-io/nominal-client/issues/580)) ([f3a3b38](https://github.com/nominal-io/nominal-client/commit/f3a3b383d62e8466af617179479918ef0d1ee534))
+
+
+### Bug Fixes
+
+* handle case where compute returns a non-bucketed response for a bucketed query ([#579](https://github.com/nominal-io/nominal-client/issues/579)) ([58b84f5](https://github.com/nominal-io/nominal-client/commit/58b84f55d282a84c8eb6703816b16920e3b8ecf3))
+
 ## [1.105.0](https://github.com/nominal-io/nominal-client/compare/v1.104.3...v1.105.0) (2026-01-21)
 
 

{nominal-1.105.0 → nominal-1.107.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: nominal
-Version: 1.105.0
+Version: 1.107.0
 Summary: Automate Nominal workflows in Python
 Project-URL: Homepage, https://nominal.io
 Project-URL: Documentation, https://docs.nominal.io
@@ -20,7 +20,7 @@ Requires-Dist: cachetools>=6.1.0
 Requires-Dist: click<9,>=8
 Requires-Dist: conjure-python-client<4,>=3.1.0
 Requires-Dist: ffmpeg-python>=0.2.0
-Requires-Dist: nominal-api==0.1032.0
+Requires-Dist: nominal-api==0.1075.0
 Requires-Dist: nominal-streaming==0.5.8; platform_python_implementation == 'CPython' and python_version >= '3.10' and ((sys_platform == 'win32' and platform_machine == 'AMD64') or (sys_platform == 'darwin' and platform_machine == 'arm64') or (sys_platform == 'linux' and (platform_machine == 'x86_64' or platform_machine == 'armv7l')))
 Requires-Dist: openpyxl>=0.0.0
 Requires-Dist: pandas>=0.0.0

{nominal-1.105.0 → nominal-1.107.0}/nominal/core/dataset.py

@@ -1093,6 +1093,7 @@ def _construct_new_ingest_options(
                 tag_columns=tag_columns,
                 is_archive=file_type.is_parquet_archive(),
                 additional_file_tags={**tags} if tags else None,
+                exclude_columns=[],
             )
         )
     else:
@@ -1107,6 +1108,7 @@ def _construct_new_ingest_options(
                 channel_prefix=channel_prefix,
                 tag_columns=tag_columns,
                 additional_file_tags={**tags} if tags else None,
+                exclude_columns=[],
             )
         )
 
@@ -1139,6 +1141,7 @@ def _construct_existing_ingest_options(
                 tag_columns=tag_columns,
                 is_archive=file_type.is_parquet_archive(),
                 additional_file_tags={**tags} if tags else None,
+                exclude_columns=[],
             )
         )
     else:
@@ -1152,5 +1155,6 @@ def _construct_existing_ingest_options(
                 timestamp_metadata=timestamp_metadata,
                 tag_columns=tag_columns,
                 additional_file_tags={**tags} if tags else None,
+                exclude_columns=[],
             )
         )

{nominal-1.105.0 → nominal-1.107.0}/nominal/experimental/compute/_buckets.py

@@ -317,10 +317,41 @@ def _compute_buckets(
 def _numeric_buckets_from_compute_response(
     response: scout_compute_api.ComputeNodeResponse,
 ) -> Iterable[tuple[api.Timestamp, scout_compute_api.NumericBucket]]:
-    if response.type != "bucketedNumeric" or response.bucketed_numeric is None:
-        return
-
-    yield from zip(response.bucketed_numeric.timestamps, response.bucketed_numeric.buckets)
+    if response.numeric_point is not None:
+        # single point would be returned-- create a synthetic bucket
+        val = response.numeric_point.value
+        yield (
+            response.numeric_point.timestamp,
+            scout_compute_api.NumericBucket(
+                count=1,
+                first_point=response.numeric_point,
+                max=val,
+                mean=val,
+                min=val,
+                variance=0,
+                last_point=response.numeric_point,
+            ),
+        )
+    elif response.numeric is not None:
+        # Not enough points to reach the number of requested bucket count, so
+        # gets returned as all of the raw data.
+        for timestamp, value in zip(response.numeric.timestamps, response.numeric.values):
+            point = scout_compute_api.NumericPoint(timestamp, value)
+            yield (
+                timestamp,
+                scout_compute_api.NumericBucket(
+                    count=1,
+                    first_point=point,
+                    max=value,
+                    min=value,
+                    mean=value,
+                    variance=0,
+                    last_point=point,
+                ),
+            )
+    elif response.bucketed_numeric is not None:
+        # Actually bucketed data
+        yield from zip(response.bucketed_numeric.timestamps, response.bucketed_numeric.buckets)
 
 
 def _timestamp_from_conjure(timestamp: api.Timestamp) -> params.NanosecondsUTC:

nominal-1.107.0/nominal/experimental/dataset_utils/__init__.py

@@ -0,0 +1,5 @@
+from nominal.experimental.dataset_utils._dataset_utils import create_dataset_with_uuid
+
+__all__ = [
+    "create_dataset_with_uuid",
+]

nominal-1.107.0/nominal/experimental/dataset_utils/_dataset_utils.py

@@ -0,0 +1,52 @@
+from collections.abc import Mapping, Sequence
+
+from nominal_api import scout_catalog
+
+from nominal.core import Dataset, NominalClient
+
+
+def create_dataset_with_uuid(
+    client: NominalClient,
+    dataset_uuid: str,
+    name: str,
+    *,
+    description: str | None = None,
+    labels: Sequence[str] = (),
+    properties: Mapping[str, str] | None = None,
+) -> Dataset:
+    """Create a dataset with a specific UUID.
+
+    This is useful for migrations where the dataset UUID must be controlled by the caller.
+    Throws a conflict error if a dataset with the specified UUID already exists.
+
+    This endpoint is not intended for general use. Use `NominalClient.create_dataset` instead
+    to create a new dataset with an auto-generated UUID.
+
+    Args:
+        client: The NominalClient to use for creating the dataset.
+        dataset_uuid: The UUID to assign to the new dataset.
+        name: Name of the dataset to create.
+        description: Human readable description of the dataset.
+        labels: Text labels to apply to the created dataset.
+        properties: Key-value properties to apply to the created dataset.
+
+    Returns:
+        Reference to the created dataset in Nominal.
+    """
+    create_dataset_request = scout_catalog.CreateDataset(
+        name=name,
+        description=description,
+        labels=list(labels),
+        properties={} if properties is None else dict(properties),
+        is_v2_dataset=True,
+        metadata={},
+        origin_metadata=scout_catalog.DatasetOriginMetadata(),
+        workspace=client._clients.workspace_rid,
+        marking_rids=[],
+    )
+    request = scout_catalog.CreateDatasetWithUuidRequest(
+        create_dataset=create_dataset_request,
+        uuid=dataset_uuid,
+    )
+    response = client._clients.catalog.create_dataset_with_uuid(client._clients.auth_header, request)
+    return Dataset._from_conjure(client._clients, response)

nominal-1.107.0/nominal/experimental/migration/migration_data_config.py

@@ -0,0 +1,11 @@
+class MigrationDatasetConfig:
+    preserve_dataset_uuid: bool
+    include_dataset_files: bool
+
+    def __init__(self, preserve_dataset_uuid: bool, include_dataset_files: bool) -> None:
+        """Args:
+        preserve_dataset_uuid (bool): If true, preserves the original dataset UUIDs during migration.
+        include_dataset_files (bool): If true, includes dataset files in the migration.
+        """
+        self.preserve_dataset_uuid = preserve_dataset_uuid
+        self.include_dataset_files = include_dataset_files

{nominal-1.105.0 → nominal-1.107.0}/nominal/experimental/migration/migration_utils.py

@@ -26,6 +26,8 @@ from nominal.core._event_types import EventType, SearchEventOriginType
 from nominal.core._utils.api_tools import Link, LinkDict
 from nominal.core.attachment import Attachment
 from nominal.core.run import Run
+from nominal.experimental.dataset_utils import create_dataset_with_uuid
+from nominal.experimental.migration.migration_data_config import MigrationDatasetConfig
 from nominal.experimental.migration.migration_resources import MigrationResources
 from nominal.ts import (
     IntegralNanosecondsDuration,
@@ -422,6 +424,7 @@ def copy_dataset_from(
     new_dataset_properties: dict[str, Any] | None = None,
     new_dataset_labels: Sequence[str] | None = None,
     include_files: bool = False,
+    preserve_uuid: bool = False,
 ) -> Dataset:
     """Copy a dataset from the source to the destination client.
 
@@ -435,6 +438,9 @@ def copy_dataset_from(
             properties are used.
         new_dataset_labels: Optional new labels for the copied dataset. If not provided, the original labels are used.
         include_files: Whether to include files in the copied dataset.
+        preserve_uuid: If True, create the dataset with the same UUID as the source dataset.
+            This is useful for migrations where references to datasets must be preserved.
+            Throws a conflict error if a dataset with the UUID already exists.
 
     Returns:
         The newly created Dataset in the destination client.
@@ -448,12 +454,34 @@ def copy_dataset_from(
         source_dataset.rid,
         extra=log_extras,
     )
-    new_dataset = destination_client.create_dataset(
-        name=new_dataset_name if new_dataset_name is not None else source_dataset.name,
-        description=new_dataset_description if new_dataset_description is not None else source_dataset.description,
-        properties=new_dataset_properties if new_dataset_properties is not None else source_dataset.properties,
-        labels=new_dataset_labels if new_dataset_labels is not None else source_dataset.labels,
-    )
+
+    dataset_name = new_dataset_name if new_dataset_name is not None else source_dataset.name
+    dataset_description = new_dataset_description if new_dataset_description is not None else source_dataset.description
+    dataset_properties = new_dataset_properties if new_dataset_properties is not None else source_dataset.properties
+    dataset_labels = new_dataset_labels if new_dataset_labels is not None else source_dataset.labels
+
+    if preserve_uuid:
+        # Extract the UUID from the source dataset's rid
+        match = UUID_PATTERN.search(source_dataset.rid)
+        if not match:
+            raise ValueError(f"Could not extract UUID from dataset rid: {source_dataset.rid}")
+        source_uuid = match.group(2)
+        new_dataset = create_dataset_with_uuid(
+            client=destination_client,
+            dataset_uuid=source_uuid,
+            name=dataset_name,
+            description=dataset_description,
+            labels=dataset_labels,
+            properties=dataset_properties,
+        )
+    else:
+        new_dataset = destination_client.create_dataset(
+            name=dataset_name,
+            description=dataset_description,
+            properties=dataset_properties,
+            labels=dataset_labels,
+        )
+
     if include_files:
         for source_file in source_dataset.list_files():
             copy_file_to_dataset(source_file, new_dataset)
@@ -612,7 +640,7 @@ def clone_asset(
     return copy_asset_from(
         source_asset=source_asset,
         destination_client=destination_client,
-        include_data=True,
+        dataset_config=MigrationDatasetConfig(preserve_dataset_uuid=True, include_dataset_files=True),
         include_events=True,
         include_runs=True,
     )
@@ -626,7 +654,7 @@ def copy_asset_from(
     new_asset_description: str | None = None,
     new_asset_properties: dict[str, Any] | None = None,
     new_asset_labels: Sequence[str] | None = None,
-    include_data: bool = False,
+    dataset_config: MigrationDatasetConfig | None = None,
     include_events: bool = False,
     include_runs: bool = False,
 ) -> Asset:
@@ -639,7 +667,7 @@ def copy_asset_from(
         new_asset_description: Optional new description for the copied asset. If not provided, original description used
         new_asset_properties: Optional new properties for the copied asset. If not provided, original properties used.
         new_asset_labels: Optional new labels for the copied asset. If not provided, the original labels are used.
-        include_data: Whether to include data in the copied asset.
+        dataset_config: Configuration for dataset migration.
         include_events: Whether to include events in the copied dataset.
         include_runs: Whether to include runs in the copied asset.
 
@@ -656,15 +684,16 @@ def copy_asset_from(
         properties=new_asset_properties if new_asset_properties is not None else source_asset.properties,
         labels=new_asset_labels if new_asset_labels is not None else source_asset.labels,
     )
-    if include_data:
+    if dataset_config is not None:
         source_datasets = source_asset.list_datasets()
         for data_scope, source_dataset in source_datasets:
-            new_dataset = clone_dataset(
+            new_dataset = copy_dataset_from(
                 source_dataset=source_dataset,
                 destination_client=destination_client,
+                preserve_uuid=dataset_config.preserve_dataset_uuid,
+                include_files=dataset_config.include_dataset_files,
             )
             new_asset.add_dataset(data_scope, new_dataset)
-    source_asset._list_dataset_scopes
 
     if include_events:
         source_events = source_asset.search_events(origin_types=SearchEventOriginType.get_manual_origin_types())
@@ -683,6 +712,7 @@ def copy_asset_from(
 def copy_resources_to_destination_client(
     destination_client: NominalClient,
     migration_resources: MigrationResources,
+    dataset_config: MigrationDatasetConfig | None = None,
 ) -> tuple[Sequence[tuple[str, Dataset]], Sequence[Asset], Sequence[WorkbookTemplate], Sequence[Workbook]]:
     """Based on a list of assets and workbook templates, copy resources to destination client, creating
        new datasets, datafiles, and workbooks along the way.
@@ -690,6 +720,7 @@ def copy_resources_to_destination_client(
     Args:
         destination_client (NominalClient): client of the tenant/workspace to copy resources to.
         migration_resources (MigrationResources): resources to copy.
+        dataset_config (MigrationDataConfig | None): Configuration for dataset migration.
 
     Returns:
         All of the created resources.
@@ -704,7 +735,13 @@ def copy_resources_to_destination_client(
 
     new_data_scopes_and_datasets: list[tuple[str, Dataset]] = []
     for source_asset in migration_resources.source_assets:
-        new_asset = clone_asset(source_asset.asset, destination_client)
+        new_asset = copy_asset_from(
+            source_asset.asset,
+            destination_client,
+            dataset_config=dataset_config,
+            include_events=True,
+            include_runs=True,
+        )
         new_assets.append(new_asset)
         new_data_scopes_and_datasets.extend(new_asset.list_datasets())
 
@@ -712,7 +749,7 @@ def copy_resources_to_destination_client(
             new_template = clone_workbook_template(source_workbook_template, destination_client)
             new_templates.append(new_template)
             new_workbook = new_template.create_workbook(
-                title=new_template.title, description=new_template.description, asset=new_assets[0]
+                title=new_template.title, description=new_template.description, asset=new_asset
             )
             logger.debug(
                 "Created new workbook %s (rid: %s) from template %s (rid: %s)",

{nominal-1.105.0 → nominal-1.107.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "nominal"
-version = "1.105.0"
+version = "1.107.0"
 description = "Automate Nominal workflows in Python"
 authors = [
     { name = "Alexander Reynolds", email = "alex.reynolds@nominal.io" },
@@ -28,7 +28,7 @@ dependencies = [
     "click>=8,<9",
     "conjure-python-client>=3.1.0,<4",
     "ffmpeg-python>=0.2.0",
-    "nominal-api==0.1032.0",
+    "nominal-api==0.1075.0",
     "rich>=14.1.0",
     "tabulate>=0.9.0,<0.10",
     "truststore>=0.10.4",