lamindb 1.6.0__tar.gz → 1.6.1__tar.gz

{lamindb-1.6.0 → lamindb-1.6.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: lamindb
-Version: 1.6.0
+Version: 1.6.1
 Summary: A data framework for biology.
 Author-email: Lamin Labs <open-source@lamin.ai>
 Requires-Python: >=3.10,<3.14
@@ -10,7 +10,7 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Requires-Dist: lamin_utils==0.14.0
-Requires-Dist: lamin_cli==1.4.1
+Requires-Dist: lamin_cli==1.4.2
 Requires-Dist: lamindb_setup[aws]==1.6.0
 Requires-Dist: pyyaml
 Requires-Dist: pyarrow
@@ -23,7 +23,7 @@ Requires-Dist: anndata>=0.8.0,<=0.11.4
 Requires-Dist: fsspec
 Requires-Dist: graphviz
 Requires-Dist: psycopg2-binary
-Requires-Dist: bionty>=1.4a1 ; extra == "bionty"
+Requires-Dist: bionty>=1.5.0 ; extra == "bionty"
 Requires-Dist: clinicore ; extra == "clinicore"
 Requires-Dist: tomlkit ; extra == "dev"
 Requires-Dist: line_profiler ; extra == "dev"
@@ -36,6 +36,7 @@ Requires-Dist: pytest-cov ; extra == "dev"
 Requires-Dist: mudata ; extra == "dev"
 Requires-Dist: nbproject_test>=0.6.0 ; extra == "dev"
 Requires-Dist: faker-biology ; extra == "dev"
+Requires-Dist: pronto ; extra == "dev"
 Requires-Dist: readfcs>=2.0.1 ; extra == "fcs"
 Requires-Dist: lamindb_setup[gcp] ; extra == "gcp"
 Requires-Dist: nbproject==0.11.1 ; extra == "jupyter"
@@ -43,7 +44,7 @@ Requires-Dist: jupytext ; extra == "jupyter"
 Requires-Dist: nbconvert>=7.2.1 ; extra == "jupyter"
 Requires-Dist: mistune!=3.1.0 ; extra == "jupyter"
 Requires-Dist: omop ; extra == "omop"
-Requires-Dist: wetlab>=1.3a1 ; extra == "wetlab"
+Requires-Dist: wetlab>=1.3.1 ; extra == "wetlab"
 Requires-Dist: numcodecs<0.16.0 ; extra == "zarr"
 Requires-Dist: zarr>=2.16.0,<3.0.0a0 ; extra == "zarr"
 Project-URL: Home, https://github.com/laminlabs/lamindb

{lamindb-1.6.0 → lamindb-1.6.1}/lamindb/__init__.py

@@ -109,7 +109,7 @@ Backwards compatibility.
 
 # ruff: noqa: I001
 # denote a release candidate for 0.1.0 with 0.1rc1, 0.1a1, 0.1b1, etc.
-__version__ = "1.6.0"
+__version__ = "1.6.1"
 
 import warnings
 
@@ -140,7 +140,6 @@ if _check_instance_setup(from_module="lamindb"):
     from .curators._legacy import CatManager as Curator
     from .models import (
         Artifact,
-        Branch,
         Collection,
         Feature,
         FeatureSet,  # backward compat
@@ -154,6 +153,7 @@ if _check_instance_setup(from_module="lamindb"):
         ULabel,
         User,
         Space,
+        Branch,
         Record,
         Sheet,
     )

{lamindb-1.6.0 → lamindb-1.6.1}/lamindb/core/_context.py

@@ -38,7 +38,7 @@ if TYPE_CHECKING:
     from lamindb_setup.core.types import UPathStr
 
     from lamindb.base.types import TransformType
-    from lamindb.models import Project, Space
+    from lamindb.models import Branch, Project, Space
 
 is_run_from_ipython = getattr(builtins, "__IPYTHON__", False)
 
@@ -193,6 +193,7 @@ class Context:
         """A local path to the script or notebook that's running."""
         self._project: Project | None = None
         self._space: Space | None = None
+        self._branch: Branch | None = None
         self._logging_message_track: str = ""
         self._logging_message_imports: str = ""
         self._stream_tracker: LogStreamTracker = LogStreamTracker()
@@ -247,9 +248,14 @@ class Context:
 
     @property
     def space(self) -> Space | None:
-        """The space in which entities are created during the run."""
+        """The space in which artifacts, collections, transforms, and runs are saved during the run."""
         return self._space
 
+    @property
+    def branch(self) -> Branch | None:
+        """The branch on which entities are created during the run."""
+        return self._branch
+
     @property
     def run(self) -> Run | None:
         """Managed run of context."""
@@ -261,6 +267,7 @@ class Context:
         *,
         project: str | Project | None = None,
         space: str | Space | None = None,
+        branch: str | Branch | None = None,
         params: dict | None = None,
         new_run: bool | None = None,
         path: str | None = None,
@@ -273,11 +280,11 @@ class Context:
 
         Args:
             transform: A transform (stem) `uid` (or record). If `None`, auto-creates a `transform` with its `uid`.
-            project: A project, its `name` or `uid` for labeling entities created during the run.
-            space: A restricted space, its `name` or `uid` for creating sensitive entities are created during the run.
-                The default is the common `"All"` space that every LaminDB instance has.
-                The `space` argument doesn't affect `Storage`, `ULabel`, `Feature`, `Schema`, `Param` and bionty entities as these provide structure that should typically be commonly accessible.
+            project: A project (or its `name` or `uid`) for labeling entities.
+            space: A restricted space (or its `name` or `uid`) in which to store artifacts, collections, transforms, and runs.
+                Default: the `"All"` space.
                 If you want to manually move entities to a different space, set the `.space` field (:doc:`docs:access`).
+            branch: A branch (or its `name` or `uid`) on which to store records.
             params: A dictionary of parameters to track for the run.
             new_run: If `False`, loads the latest run of transform
                 (default notebook), if `True`, creates new run (default non-notebook).
@@ -299,7 +306,7 @@ class Context:
 
             More examples: :doc:`/track`
         """
-        from lamindb.models import Project, Space
+        from lamindb.models import Branch, Project, Space
 
         instance_settings = ln_setup.settings.instance
         # similar logic here: https://github.com/laminlabs/lamindb/pull/2527
@@ -337,6 +344,21 @@ class Context:
                         f"Space '{space}', please check on the hub UI whether you have the correct `uid` or `name`."
                     )
             self._space = space_record
+        if branch is not None:
+            if isinstance(branch, Branch):
+                assert branch._state.adding is False, (  # noqa: S101
+                    "Branch must be saved before passing it to track()"
+                )
+                branch_record = branch
+            else:
+                branch_record = Branch.filter(
+                    Q(name=branch) | Q(uid=branch)
+                ).one_or_none()
+                if branch_record is None:
+                    raise InvalidArgument(
+                        f"Space '{branch}', please check on the hub UI whether you have the correct `uid` or `name`."
+                    )
+            self._branch = branch_record
         self._logging_message_track = ""
         self._logging_message_imports = ""
         if transform is not None and isinstance(transform, str):

{lamindb-1.6.0 → lamindb-1.6.1}/lamindb/core/datasets/mini_immuno.py

@@ -105,9 +105,9 @@ def get_dataset1(
         "treatment_time_h": [24, 24, 6],
         "donor": ["D0001", "D0002", None],
         "donor_ethnicity": [
-            ["African", "African American"],
-            ["African", "West African"],
-            ["Asian"],
+            ["Chinese", "Singaporean Chinese"],
+            ["Chinese", "Han Chinese"],
+            ["Chinese"],
         ],
     }
     # define the dataset-level metadata

{lamindb-1.6.0 → lamindb-1.6.1}/lamindb/migrations/0101_alter_artifact_hash_alter_feature_name_and_more.py

@@ -8,42 +8,35 @@ import lamindb.base.fields
 
 
 def populate_hash_values(apps, schema_editor):
+    from tqdm import tqdm
+
     FeatureValue = apps.get_model("lamindb", "FeatureValue")
 
-    # Get records that need updating
-    records_to_update = FeatureValue.objects.filter(hash__isnull=True)
+    records_queryset = FeatureValue.objects.filter(hash__isnull=True)
+    total_count = records_queryset.count()
+    print(f"\nFound {total_count} FeatureValue records with hash=None. Populating...")
 
-    print(
-        f"\nFound {records_to_update.count()} FeatureValue records with hash=None. Populating..."
-    )
+    # Convert to list to avoid QuerySet modification issues
+    records_to_update = list(records_queryset)
 
+    batch_size = 1000
     updated_count = 0
-    skipped_count = 0
 
-    # Iterate efficiently through records
-    for record in records_to_update.iterator():
-        value = record.value
-        new_hash = None
+    for batch_start in tqdm(range(0, len(records_to_update), batch_size)):
+        batch_records = records_to_update[batch_start : batch_start + batch_size]
 
-        if not isinstance(value, dict):
-            new_hash = hash_string(str(value))
-        else:
-            new_hash = hash_dict(value)
+        # Calculate hashes for all records in batch
+        for record in batch_records:
+            if not isinstance(record.value, dict):
+                record.hash = hash_string(str(record.value))
+            else:
+                record.hash = hash_dict(record.value)
 
-        try:
-            record.hash = new_hash
-            record.save(update_fields=["hash"])
-            updated_count += 1
-        except Exception as e:
-            # This will likely catch IntegrityError if a duplicate hash exists
-            # for the same feature. You might want to log these or handle them
-            # based on your data cleanup strategy.
-            print(
-                f"  - Could not update record {record.id} (value: {str(value)[:50]}...): {e}"
-            )
-            skipped_count += 1
+        # Bulk update the entire batch
+        FeatureValue.objects.bulk_update(batch_records, ["hash"], batch_size=batch_size)
+        updated_count += len(batch_records)
 
-    print(f"Finished. Updated: {updated_count}, Skipped/Errors: {skipped_count}.")
+    print(f"Finished. Updated: {updated_count} records.")
 
 
 class Migration(migrations.Migration):

lamindb-1.6.1/lamindb/migrations/0104_alter_branch_uid.py

@@ -0,0 +1,403 @@
+# Generated by Django 5.2 on 2025-06-02 10:53
+
+import django.db.models.deletion
+from django.db import migrations
+
+import lamindb.base.fields
+import lamindb.base.uids
+
+
+class Migration(migrations.Migration):
+    dependencies = [
+        ("lamindb", "0103_remove_writelog_migration_state_and_more"),
+    ]
+
+    operations = [
+        migrations.AlterField(
+            model_name="branch",
+            name="uid",
+            field=lamindb.base.fields.CharField(
+                blank=True,
+                db_index=True,
+                default=lamindb.base.uids.base62_12,
+                editable=False,
+                max_length=12,
+                unique=True,
+            ),
+        ),
+        migrations.AlterField(
+            model_name="artifact",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="artifact",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="collection",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="collection",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="feature",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="feature",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="featurevalue",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="featurevalue",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="person",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="person",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="project",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="project",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="record",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="record",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="recordrecord",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="recordrecord",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="reference",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="reference",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="run",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="run",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="schema",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="schema",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="sheet",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="sheet",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="storage",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="storage",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="transform",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="transform",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="ulabel",
+            name="branch",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_column="_branch_code",
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.branch",
+            ),
+        ),
+        migrations.AlterField(
+            model_name="ulabel",
+            name="space",
+            field=lamindb.base.fields.ForeignKey(
+                blank=True,
+                db_default=1,
+                default=1,
+                on_delete=django.db.models.deletion.PROTECT,
+                related_name="+",
+                to="lamindb.space",
+            ),
+        ),
+    ]