datachain 0.20.3__tar.gz → 0.20.4__tar.gz

{datachain-0.20.3 → datachain-0.20.4}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.20.3
+Version: 0.20.4
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.20.3 → datachain-0.20.4}/src/datachain/data_storage/metastore.py

@@ -176,7 +176,9 @@ class AbstractMetastore(ABC, Serializable):
 
     @cached_property
     def default_project(self) -> Project:
-        return self.get_project(self.default_project_name, self.default_namespace_name)
+        return self.get_project(
+            self.default_project_name, self.default_namespace_name, create=True
+        )
 
     @cached_property
     def listing_project(self) -> Project:

{datachain-0.20.3 → datachain-0.20.4}/src/datachain/data_storage/sqlite.py

@@ -471,9 +471,6 @@ class SQLiteMetastore(AbstractDBMetastore):
         system_namespace = self.create_namespace(Namespace.system(), "System namespace")
         self.create_project(system_namespace.name, Project.listing(), "Listing project")
 
-        local_namespace = self.create_namespace(Namespace.default(), "Local namespace")
-        self.create_project(local_namespace.name, Project.default(), "Local project")
-
     def _check_schema_version(self) -> None:
         """
         Checks if current DB schema is up to date with latest DB model and schema

{datachain-0.20.3 → datachain-0.20.4}/src/datachain/lib/dc/datachain.py

@@ -38,7 +38,6 @@ from datachain.lib.file import (
     FileExporter,
 )
 from datachain.lib.file import ExportPlacement as FileExportPlacement
-from datachain.lib.projects import get as get_project
 from datachain.lib.settings import Settings
 from datachain.lib.signal_schema import SignalSchema
 from datachain.lib.udf import Aggregator, BatchMapper, Generator, Mapper, UDFBase
@@ -525,8 +524,10 @@ class DataChain:
         It returns the chain itself.
         """
         schema = self.signals_schema.clone_without_sys_signals().serialize()
-        project = get_project(
-            self.project_name, self.namespace_name, session=self.session
+        project = self.session.catalog.metastore.get_project(
+            self.project_name,
+            self.namespace_name,
+            create=True,
         )
         return self._evolve(
             query=self._query.save(project=project, feature_schema=schema)

{datachain-0.20.3 → datachain-0.20.4}/src/datachain/lib/dc/datasets.py

@@ -2,7 +2,11 @@ from collections.abc import Sequence
 from typing import TYPE_CHECKING, Optional, Union, get_origin, get_type_hints
 
 from datachain.dataset import parse_dataset_name
-from datachain.error import DatasetVersionNotFoundError
+from datachain.error import (
+    DatasetNotFoundError,
+    DatasetVersionNotFoundError,
+    ProjectNotFoundError,
+)
 from datachain.lib.dataset_info import DatasetInfo
 from datachain.lib.file import (
     File,
@@ -136,9 +140,15 @@ def read_dataset(
             # all 2.* dataset versions). If dataset doesn't have any versions where
             # major part is equal to that input, exception is thrown.
             major = int(version)
-            dataset = session.catalog.get_dataset(
-                name, get_project(project_name, namespace_name, session=session)
-            )
+            try:
+                ds_project = get_project(project_name, namespace_name, session=session)
+            except ProjectNotFoundError:
+                raise DatasetNotFoundError(
+                    f"Dataset {name} not found in namespace {namespace_name} and",
+                    f" project {project_name}",
+                ) from None
+
+            dataset = session.catalog.get_dataset(name, ds_project)
             latest_major = dataset.latest_major_version(major)
             if not latest_major:
                 raise DatasetVersionNotFoundError(
@@ -321,7 +331,13 @@ def delete_dataset(
             None, name, namespace_name, project_name, version=version, force=force
         )
 
-    ds_project = get_project(project_name, namespace_name, session=session)
+    try:
+        ds_project = get_project(project_name, namespace_name, session=session)
+    except ProjectNotFoundError:
+        raise DatasetNotFoundError(
+            f"Dataset {name} not found in namespace {namespace_name} and project",
+            f" {project_name}",
+        ) from None
 
     if not force:
         version = version or catalog.get_dataset(name, ds_project).latest_version

{datachain-0.20.3 → datachain-0.20.4}/src/datachain.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: datachain
-Version: 0.20.3
+Version: 0.20.4
 Summary: Wrangle unstructured AI data at scale
 Author-email: Dmitry Petrov <support@dvc.org>
 License-Expression: Apache-2.0

{datachain-0.20.3 → datachain-0.20.4}/tests/unit/lib/test_namespace.py

@@ -61,14 +61,7 @@ def test_get_namespace_not_found(test_session, dev_namespace):
     assert str(excinfo.value) == "Namespace wrong not found."
 
 
-def test_local_namespace_is_created(test_session):
-    namespace_class = test_session.catalog.metastore.namespace_class
-    local_namespace = get_namespace(namespace_class.default(), session=test_session)
-    assert local_namespace.name == namespace_class.default()
-
-
 def test_ls_namespaces(test_session):
-    default_namespace_name = test_session.catalog.metastore.default_namespace_name
     system_namespace_name = test_session.catalog.metastore.system_namespace_name
 
     create_namespace("ns1")
@@ -76,14 +69,11 @@ def test_ls_namespaces(test_session):
 
     namespaces = ls_namespaces(session=test_session)
     assert sorted([n.name for n in namespaces]) == sorted(
-        [default_namespace_name, system_namespace_name, "ns1", "ns2"]
+        [system_namespace_name, "ns1", "ns2"]
     )
 
 
 def test_ls_namespaces_just_local(test_session):
-    default_namespace_name = test_session.catalog.metastore.default_namespace_name
     system_namespace_name = test_session.catalog.metastore.system_namespace_name
     namespaces = ls_namespaces(session=test_session)
-    assert sorted([n.name for n in namespaces]) == sorted(
-        [default_namespace_name, system_namespace_name]
-    )
+    assert [n.name for n in namespaces] == [system_namespace_name]

{datachain-0.20.3 → datachain-0.20.4}/tests/unit/lib/test_project.py

@@ -102,16 +102,6 @@ def test_get_project_not_found_but_exists_in_other_namespace(
     assert str(excinfo.value) == f"Project {name} in namespace dev not found."
 
 
-@skip_if_not_sqlite
-def test_local_project_is_created(test_session):
-    project_class = test_session.catalog.metastore.project_class
-    namespace_class = test_session.catalog.metastore.namespace_class
-    local_project = get_project(
-        project_class.default(), namespace_class.default(), session=test_session
-    )
-    assert local_project.name == project_class.default()
-
-
 def test_ls_projects(test_session):
     metastore = test_session.catalog.metastore
 
@@ -123,7 +113,6 @@ def test_ls_projects(test_session):
     projects = ls_projects(session=test_session)
     assert sorted([(p.namespace.name, p.name) for p in projects]) == sorted(
         [
-            (metastore.default_namespace_name, metastore.default_project_name),
             (metastore.system_namespace_name, metastore.listing_project_name),
             ("ns1", "p1"),
             ("ns1", "p2"),
@@ -157,7 +146,6 @@ def test_ls_projects_just_default(test_session):
     projects = ls_projects(session=test_session)
     assert sorted([(p.namespace.name, p.name) for p in projects]) == sorted(
         [
-            (metastore.default_namespace_name, metastore.default_project_name),
             (metastore.system_namespace_name, metastore.listing_project_name),
         ]
     )