Merge branch 'develop' of https://github.com/Labelbox/labelbox-python into ms/al-3814

Author: Matt Sokoloff

docs/source/index.rst

@@ -173,14 +173,14 @@ DataRowMetadata
 
 .. automodule:: labelbox.schema.data_row_metadata
    :members:
-   :show-inheritance:   
+   :show-inheritance:
 
 AnnotationImport
 ----------------------------
 
 .. automodule:: labelbox.schema.annotation_import
    :members:
-   :show-inheritance:      
+   :show-inheritance:
 
 Batch
 ----------------------------
@@ -194,4 +194,19 @@ ResourceTag
 
 .. automodule:: labelbox.schema.resource_tag
    :members:
-   :show-inheritance:
+   :show-inheritance:
+
+Slice
+-----------------------------------------
+
+.. automodule:: labelbox.schema.slice
+   :members: Slice
+   :exclude-members: CatalogSlice
+   :show-inheritance:
+
+CatalogSlice
+-----------------------------------------
+.. automodule:: labelbox.schema.slice
+   :members: CatalogSlice
+   :exclude-members: Slice
+   :show-inheritance:

labelbox/__init__.py

@@ -27,3 +27,4 @@
 from labelbox.schema.resource_tag import ResourceTag
 from labelbox.schema.project_resource_tag import ProjectResourceTag
 from labelbox.schema.media_type import MediaType
+from labelbox.schema.slice import Slice, CatalogSlice

labelbox/client.py

@@ -33,6 +33,7 @@
 from labelbox.schema.user import User
 from labelbox.schema.project import Project
 from labelbox.schema.role import Role
+from labelbox.schema.slice import CatalogSlice
 
 from labelbox.schema.media_type import MediaType
 
@@ -963,7 +964,7 @@ def assign_global_keys_to_data_rows(
             timeout_seconds=60) -> Dict[str, Union[str, List[Any]]]:
         """
         Assigns global keys to data rows.
-        
+
         Args:
             A list of dicts containing data_row_id and global_key.
         Returns:
@@ -1211,3 +1212,27 @@ def _format_failed_rows(rows: List[str],
                     "Timed out waiting for get_data_rows_for_global_keys job to complete."
                 )
             time.sleep(sleep_time)
+
+    def get_catalog_slice(self, slice_id) -> CatalogSlice:
+        """
+        Fetches a Catalog Slice by ID.
+
+        Args:
+             slice_id (str): The ID of the Slice
+        Returns:
+            CatalogSlice
+        """
+        query_str = """
+            query getSavedQueryPyApi($id: ID!) {
+                getSavedQuery(id: $id) {
+                    id
+                    name
+                    description
+                    filter
+                    createdAt
+                    updatedAt
+                }
+            }
+        """
+        res = self.execute(query_str, {'id': slice_id})
+        return Entity.CatalogSlice(self, res['getSavedQuery'])

labelbox/orm/model.py

@@ -377,6 +377,7 @@ class Entity(metaclass=EntityMeta):
         ProjectRole: Type[labelbox.ProjectRole]
         Project: Type[labelbox.Project]
         Batch: Type[labelbox.Batch]
+        CatalogSlice: Type[labelbox.CatalogSlice]
 
     @classmethod
     def _attributes_of_type(cls, attr_type):

labelbox/schema/slice.py

@@ -0,0 +1,59 @@
+from labelbox.orm.db_object import DbObject
+from labelbox.orm.model import Field
+from labelbox.pagination import PaginatedCollection
+
+
+class Slice(DbObject):
+    """
+    A Slice is a saved set of filters (saved query).
+    This is an abstract class and should not be instantiated.
+
+    Attributes:
+        name (datetime)
+        description (datetime)
+        created_at (datetime)
+        updated_at (datetime)
+        filter (json)
+    """
+    name = Field.String("name")
+    description = Field.String("description")
+    created_at = Field.DateTime("created_at")
+    updated_at = Field.DateTime("updated_at")
+    filter = Field.Json("filter")
+
+
+class CatalogSlice(Slice):
+    """
+    Represents a Slice used for filtering data rows in Catalog.
+    """
+
+    def get_data_row_ids(self) -> PaginatedCollection:
+        """
+        Fetches all data row ids that match this Slice
+
+        Returns:
+            A PaginatedCollection of data row ids
+        """
+        query_str = """
+            query getDataRowIdsBySavedQueryPyApi($id: ID!, $from: String, $first: Int!) {
+                getDataRowIdsBySavedQuery(input: {
+                    savedQueryId: $id,
+                    after: $from
+                    first: $first
+                }) {
+                    totalCount
+                    nodes
+                    pageInfo {
+                        endCursor
+                        hasNextPage
+                    }
+                }
+            }
+        """
+        return PaginatedCollection(
+            client=self.client,
+            query=query_str,
+            params={'id': self.uid},
+            dereferencing=['getDataRowIdsBySavedQuery', 'nodes'],
+            obj_class=lambda _, data_row_id: data_row_id,
+            cursor_path=['getDataRowIdsBySavedQuery', 'pageInfo', 'endCursor'])

tests/integration/conftest.py

@@ -15,6 +15,7 @@
 from labelbox.orm import query
 from labelbox.pagination import PaginatedCollection
 from labelbox.schema.annotation_import import LabelImport
+from labelbox.schema.enums import AnnotationImportState
 from labelbox.schema.invite import Invite
 from labelbox.schema.queue_mode import QueueMode
 from labelbox.schema.user import User
@@ -334,6 +335,7 @@ def create_label():
         upload_task = LabelImport.create_from_objects(
             client, project.uid, f'label-import-{uuid.uuid4()}', predictions)
         upload_task.wait_until_done(sleep_time_seconds=5)
+        assert upload_task.state == AnnotationImportState.FINISHED
 
     project.create_label = create_label
     project.create_label()

tests/integration/test_data_rows.py

@@ -151,14 +151,14 @@ def test_data_row_bulk_creation(dataset, rand_gen, image_url):
 @pytest.mark.slow
 def test_data_row_large_bulk_creation(dataset, image_url):
     # Do a longer task and expect it not to be complete immediately
-    n_local = 2000
-    n_urls = 250
+    n_urls = 1000
+    n_local = 250
     with NamedTemporaryFile() as fp:
         fp.write("Test data".encode())
         fp.flush()
         task = dataset.create_data_rows([{
             DataRow.row_data: image_url
-        }] * n_local + [fp.name] * n_urls)
+        }] * n_urls + [fp.name] * n_local)
     task.wait_till_done()
     assert task.status == "COMPLETE"
     assert len(list(dataset.data_rows())) == n_local + n_urls
@@ -353,7 +353,7 @@ def test_create_data_rows_with_invalid_metadata(dataset, image_url):
         DataRow.metadata_fields: fields
     }])
     task.wait_till_done()
-    assert task.status == "COMPLETE"
+    assert task.status == "FAILED"
     assert len(task.failed_data_rows) > 0
 
 
@@ -634,9 +634,10 @@ def test_data_row_bulk_creation_with_same_global_keys(dataset, sample_image):
     }])
 
     task.wait_till_done()
-    assert task.status == "COMPLETE"
+    assert task.status == "FAILED"
     assert len(task.failed_data_rows) > 0
     assert len(list(dataset.data_rows())) == 0
+    assert task.errors == "Import job failed"
 
     task = dataset.create_data_rows([{
         DataRow.row_data: sample_image,

tests/integration/test_delegated_access.py

@@ -6,7 +6,7 @@
 from labelbox import Client
 
 
-@pytest.mark.skipif(os.environ.get("DA_GCP_LABELBOX_API_KEY") is None,
+@pytest.mark.skipif(not os.environ.get('DA_GCP_LABELBOX_API_KEY'),
                     reason="DA_GCP_LABELBOX_API_KEY not found")
 def test_default_integration():
     """
@@ -28,7 +28,7 @@ def test_default_integration():
     ds.delete()
 
 
-@pytest.mark.skipif(os.environ.get("DA_GCP_LABELBOX_API_KEY") is None,
+@pytest.mark.skipif(not os.environ.get("DA_GCP_LABELBOX_API_KEY"),
                     reason="DA_GCP_LABELBOX_API_KEY not found")
 def test_non_default_integration():
     """

tests/integration/test_label.py

@@ -46,6 +46,7 @@ def test_label_export(configured_project_with_label):
 # TODO: Skipping this test in staging due to label not updating
 @pytest.mark.skipif(condition=os.environ['LABELBOX_TEST_ENVIRON'] == "onprem" or
                     os.environ['LABELBOX_TEST_ENVIRON'] == "staging" or
+                    os.environ['LABELBOX_TEST_ENVIRON'] == "local" or
                     os.environ['LABELBOX_TEST_ENVIRON'] == "custom",
                     reason="does not work for onprem")
 def test_label_update(configured_project_with_label):

tests/integration/test_task.py

@@ -1,6 +1,6 @@
 import pytest
 
-from labelbox import DataRow, Task
+from labelbox import DataRow
 from labelbox.schema.data_row_metadata import DataRowMetadataField
 
 EMBEDDING_SCHEMA_ID = "ckpyije740000yxdk81pbgjdc"
@@ -21,13 +21,12 @@ def test_task_errors(dataset, image_url):
             ]
         },
     ])
+
     assert task in client.get_user().created_tasks()
     task.wait_till_done()
-    assert task.status == "COMPLETE"
+    assert task.status == "FAILED"
     assert len(task.failed_data_rows) > 0
     assert task.errors is not None
-    assert 'message' in task.errors[0]
-    assert len(task.result) == 0
 
 
 def test_task_success_json(dataset, image_url):