cognite-neat 0.119.2__tar.gz → 0.119.3__tar.gz

{cognite_neat-0.119.2 → cognite_neat-0.119.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: cognite-neat
-Version: 0.119.2
+Version: 0.119.3
 Summary: Knowledge graph transformation
 License: Apache-2.0
 Author: Nikola Vasiljevic
@@ -80,7 +80,42 @@ NEAT is a domain expert centric and developer friendly solution for rapid:
 NEAT is using open and globally recognized standards maintained by the [World Wide Web Consortium (W3C)](https://www.w3.org/RDF/).
 NEAT represents an essential tool for creation of standardized, machine-actionable, linked and semantic (meta)data.
 
-> NEAT is an acronym derived from k**N**owl**Ed**ge gr**A**ph **T**ransformer hallucinated by GenAI.
+> NEAT is a funny acronym derived from k**N**owl**Ed**ge gr**A**ph **T**ransformer produced using [ACRONIMIFY](https://acronymify.com/NEAT/?q=knowledge+graph+transformer).
+
+
+## History
+
+NEAT emerged from years of experience in semantic tooling and knowledge graph development. The foundation was laid in 2020 with [sheet2rdf](https://github.com/nikokaoja/sheet2rdf), an Excel-based tool that trained data stewards to build domain-specific knowledge graphs and supported CI/CD processes in the Dutch Covid program and european wind energy community.
+
+By mid of 2022, sheet2rdf was used in several POCs in Cognite. As Cognite's Data Modeling Service (DMS) development progressed, the need for simplified data modeling experience led to demonstration of proto-NEAT, known as [sheet2fdm](https://github.com/cognitedata/sheet2fdm), an extension of sheet2rdf, enabling semantic data model definitions in OWL, SHACL, Python and GraphQL (see e.g., [wind energy data model](https://cognitedata.github.io/wind-energy-data-model/)) using a simplified version of sheet2rdf Excel template.
+
+Presented in various forums in 2022, this approach paved the way for NEAT’s formal development in November 2022 to enable cost-saving and empowerment of Cognite customers to self-sufficiently maintain and onboard knowledge graphs to Cognite Data Fusion.
+
+
+## Authorship
+
+### Authors
+The plot below shows the NEAT authorship from the start until present day.
+
+![NEAT authorship](./docs/artifacts/figs/authorship.png)
+
+#### Current authors
+- [Nikola Vasiljević](www.linkedin.com/in/thisisnikola)
+- [Anders Albert](https://www.linkedin.com/in/anders-albert-00790483/)
+- [Rogerio Júnior](https://www.linkedin.com/in/rogerio-saboia-j%C3%BAnior-087118a7/)
+
+#### Former authors
+- [Aleksandrs Livincovs](https://www.linkedin.com/in/aleksandrslivincovs/)
+- [Julia Graham](https://www.linkedin.com/in/julia-graham-959a78a7/)
+
+### Contributors
+We are very grateful for the contributions made by:
+
+
+- [Marie Solvik Lepoutre](https://www.linkedin.com/in/mslepoutre/), who improved RDF triples projections to Cognite Data Fusion
+- [Bård Henning Tvedt](https://www.linkedin.com/in/bhtvedt/), who implemented IMF importer
+- [Kristina Tomičić](https://www.linkedin.com/in/kristina-tomicic-6bb443108/), who implemented Data Model and Instances visualization
+
 
 ## Installation
 

{cognite_neat-0.119.2 → cognite_neat-0.119.3}/README.md

@@ -20,7 +20,42 @@ NEAT is a domain expert centric and developer friendly solution for rapid:
 NEAT is using open and globally recognized standards maintained by the [World Wide Web Consortium (W3C)](https://www.w3.org/RDF/).
 NEAT represents an essential tool for creation of standardized, machine-actionable, linked and semantic (meta)data.
 
-> NEAT is an acronym derived from k**N**owl**Ed**ge gr**A**ph **T**ransformer hallucinated by GenAI.
+> NEAT is a funny acronym derived from k**N**owl**Ed**ge gr**A**ph **T**ransformer produced using [ACRONIMIFY](https://acronymify.com/NEAT/?q=knowledge+graph+transformer).
+
+
+## History
+
+NEAT emerged from years of experience in semantic tooling and knowledge graph development. The foundation was laid in 2020 with [sheet2rdf](https://github.com/nikokaoja/sheet2rdf), an Excel-based tool that trained data stewards to build domain-specific knowledge graphs and supported CI/CD processes in the Dutch Covid program and european wind energy community.
+
+By mid of 2022, sheet2rdf was used in several POCs in Cognite. As Cognite's Data Modeling Service (DMS) development progressed, the need for simplified data modeling experience led to demonstration of proto-NEAT, known as [sheet2fdm](https://github.com/cognitedata/sheet2fdm), an extension of sheet2rdf, enabling semantic data model definitions in OWL, SHACL, Python and GraphQL (see e.g., [wind energy data model](https://cognitedata.github.io/wind-energy-data-model/)) using a simplified version of sheet2rdf Excel template.
+
+Presented in various forums in 2022, this approach paved the way for NEAT’s formal development in November 2022 to enable cost-saving and empowerment of Cognite customers to self-sufficiently maintain and onboard knowledge graphs to Cognite Data Fusion.
+
+
+## Authorship
+
+### Authors
+The plot below shows the NEAT authorship from the start until present day.
+
+![NEAT authorship](./docs/artifacts/figs/authorship.png)
+
+#### Current authors
+- [Nikola Vasiljević](www.linkedin.com/in/thisisnikola)
+- [Anders Albert](https://www.linkedin.com/in/anders-albert-00790483/)
+- [Rogerio Júnior](https://www.linkedin.com/in/rogerio-saboia-j%C3%BAnior-087118a7/)
+
+#### Former authors
+- [Aleksandrs Livincovs](https://www.linkedin.com/in/aleksandrslivincovs/)
+- [Julia Graham](https://www.linkedin.com/in/julia-graham-959a78a7/)
+
+### Contributors
+We are very grateful for the contributions made by:
+
+
+- [Marie Solvik Lepoutre](https://www.linkedin.com/in/mslepoutre/), who improved RDF triples projections to Cognite Data Fusion
+- [Bård Henning Tvedt](https://www.linkedin.com/in/bhtvedt/), who implemented IMF importer
+- [Kristina Tomičić](https://www.linkedin.com/in/kristina-tomicic-6bb443108/), who implemented Data Model and Instances visualization
+
 
 ## Installation
 

{cognite_neat-0.119.2 → cognite_neat-0.119.3}/cognite/neat/_constants.py

@@ -65,76 +65,40 @@ COGNITE_SPACES = frozenset(
     }
 )
 
-COGNITE_CORE_CONCEPTS = frozenset(
-    {
-        "CogniteFile",
-        "CogniteCubeMap",
-        "CogniteCADRevision",
-        "CognitePointCloudVolume",
-        "Cognite360ImageAnnotation",
-        "CogniteAnnotation",
-        "CogniteUnit",
-        "CogniteAsset",
-        "Cognite3DObject",
-        "Cognite3DRevision",
-        "Cognite360Image",
-        "CogniteDiagramAnnotation",
-        "Cognite360ImageCollection",
-        "Cognite360ImageStation",
-        "CognitePointCloudModel",
-        "CogniteTimeSeries",
-        "Cognite3DTransformation",
-        "CogniteEquipment",
-        "Cognite360ImageModel",
-        "CogniteAssetClass",
-        "CogniteAssetType",
-        "CogniteEquipmentType",
-        "Cognite3DModel",
-        "CogniteCADModel",
-        "CognitePointCloudRevision",
-        "CogniteCADNode",
-        "CogniteFileCategory",
-        "CogniteActivity",
-    }
-)
-
-
-COGNITE_CORE_FEATURES = frozenset(
-    {
-        "CogniteDescribable",
-        "CogniteSourceable",
-        "CogniteSourceSystem",
-        "CogniteSchedulable",
-        "CogniteVisualizable",
-    }
-)
-
-COGNITE_3D_CONCEPTS = frozenset(
-    {
-        "Cognite3DModel",
-        "Cognite3DObject",
-        "Cognite3DRevision",
-        "Cognite3DTransformation",
-        "Cognite360Image",
-        "Cognite360ImageAnnotation",
-        "Cognite360ImageCollection",
-        "Cognite360ImageModel",
-        "Cognite360ImageStation",
-        "CogniteCADModel",
-        "CogniteCADNode",
-        "CogniteCADRevision",
-        "CogniteCubeMap",
-        "CognitePointCloudModel",
-        "CognitePointCloudRevision",
-        "CognitePointCloudVolume",
-    }
-)
-
-COGNITE_ANNOTATION = frozenset(
-    {
-        "CogniteAnnotation",
-        "CogniteDiagramAnnotation",
-    }
+COGNITE_CONCEPTS = (
+    "CogniteAsset",
+    "CogniteEquipment",
+    "CogniteActivity",
+    "CogniteTimeSeries",
+    "CogniteFile",
+    "CogniteUnit",
+    "CogniteAssetClass",
+    "CogniteAssetType",
+    "CogniteEquipmentType",
+    "CogniteFileCategory",
+    "CogniteDescribable",
+    "CogniteSourceable",
+    "CogniteSourceSystem",
+    "CogniteSchedulable",
+    "CogniteVisualizable",
+    "CogniteAnnotation",
+    "CogniteDiagramAnnotation",
+    "CogniteCubeMap",
+    "CogniteCADRevision",
+    "CognitePointCloudVolume",
+    "Cognite360ImageAnnotation",
+    "Cognite3DObject",
+    "Cognite3DRevision",
+    "Cognite360Image",
+    "Cognite360ImageCollection",
+    "Cognite360ImageStation",
+    "CognitePointCloudModel",
+    "Cognite3DTransformation",
+    "Cognite360ImageModel",
+    "Cognite3DModel",
+    "CogniteCADModel",
+    "CognitePointCloudRevision",
+    "CogniteCADNode",
 )
 
 DMS_LISTABLE_PROPERTY_LIMIT = 1000

{cognite_neat-0.119.2 → cognite_neat-0.119.3}/cognite/neat/_graph/extractors/__init__.py

@@ -10,11 +10,9 @@ from ._classic_cdf._labels import LabelsExtractor
 from ._classic_cdf._relationships import RelationshipsExtractor
 from ._classic_cdf._sequences import SequencesExtractor
 from ._classic_cdf._timeseries import TimeSeriesExtractor
-from ._dexpi import DexpiExtractor
 from ._dict import DictExtractor
 from ._dms import DMSExtractor
 from ._dms_graph import DMSGraphExtractor
-from ._iodd import IODDExtractor
 from ._mock_graph_generator import MockGraphGenerator
 from ._raw import RAWExtractor
 from ._rdf_file import RdfFileExtractor
@@ -26,11 +24,9 @@ __all__ = [
     "DMSExtractor",
     "DMSGraphExtractor",
     "DataSetExtractor",
-    "DexpiExtractor",
     "DictExtractor",
     "EventsExtractor",
     "FilesExtractor",
-    "IODDExtractor",
     "KnowledgeGraphExtractor",
     "LabelsExtractor",
     "MockGraphGenerator",
@@ -52,8 +48,6 @@ TripleExtractors = (
     | FilesExtractor
     | LabelsExtractor
     | RdfFileExtractor
-    | DexpiExtractor
-    | IODDExtractor
     | DMSExtractor
     | ClassicGraphExtractor
     | DataSetExtractor

{cognite_neat-0.119.2 → cognite_neat-0.119.3}/cognite/neat/_graph/loaders/_rdf2dms.py

@@ -259,7 +259,7 @@ class DMSLoader(CDFLoader[dm.InstanceApply]):
         """Selects the views with data."""
         view_iterations: dict[dm.ViewId, _ViewIterator] = {}
         for view_id, query in view_query_by_id.items():
-            count = self.graph_store.queries.count_of_type(query.rdf_type)
+            count = self.graph_store.queries.select.count_of_type(query.rdf_type)
             if count > 0:
                 view_iterations[view_id] = _ViewIterator(view_id, count, query)
         return view_iterations
@@ -269,7 +269,7 @@ class DMSLoader(CDFLoader[dm.InstanceApply]):
         if self._space_property is None:
             return issues
         total = sum(it.instance_count for it in view_iterations)
-        properties_by_uriref = self.graph_store.queries.properties()
+        properties_by_uriref = self.graph_store.queries.select.properties()
         space_property_uri = next((k for k, v in properties_by_uriref.items() if v == self._space_property), None)
         if space_property_uri is None:
             error: ResourceNotFoundError[str, str] = ResourceNotFoundError(
@@ -282,7 +282,7 @@ class DMSLoader(CDFLoader[dm.InstanceApply]):
             issues.append(error)
             return issues
 
-        instance_iterable = self.graph_store.queries.list_instances_ids_by_space(space_property_uri)
+        instance_iterable = self.graph_store.queries.select.list_instances_ids_by_space(space_property_uri)
         instance_iterable = iterate_progress_bar_if_above_config_threshold(
             instance_iterable, total, f"Looking up spaces for {total} instances..."
         )
@@ -308,8 +308,8 @@ class DMSLoader(CDFLoader[dm.InstanceApply]):
         if not self.neat_prefix_by_type_uri:
             return
 
-        count = sum(count for _, count in self.graph_store.queries.summarize_instances())
-        instance_iterable = self.graph_store.queries.list_instances_ids()
+        count = sum(count for _, count in self.graph_store.queries.select.summarize_instances())
+        instance_iterable = self.graph_store.queries.select.list_instances_ids()
         instance_iterable = iterate_progress_bar_if_above_config_threshold(
             instance_iterable, count, f"Looking up identifiers for {count} instances..."
         )

cognite_neat-0.119.3/cognite/neat/_graph/queries/__init__.py

@@ -0,0 +1,3 @@
+from ._queries import Queries
+
+__all__ = ["Queries"]

cognite_neat-0.119.3/cognite/neat/_graph/queries/_base.py

@@ -0,0 +1,16 @@
+from rdflib import Dataset, Graph, URIRef
+from rdflib.graph import DATASET_DEFAULT_GRAPH_ID
+
+
+class BaseQuery:
+    def __init__(
+        self,
+        dataset: Dataset,
+        default_named_graph: URIRef | None = None,
+    ):
+        self.dataset = dataset
+        self.default_named_graph = default_named_graph or DATASET_DEFAULT_GRAPH_ID
+
+    def graph(self, named_graph: URIRef | None = None) -> Graph:
+        """Get named graph from the dataset to query over"""
+        return self.dataset.graph(named_graph or self.default_named_graph)

cognite_neat-0.119.3/cognite/neat/_graph/queries/_queries.py

@@ -0,0 +1,16 @@
+from rdflib import Dataset, URIRef
+
+from ._select import SelectQueries
+from ._update import UpdateQueries
+
+
+class Queries:
+    """Helper class for storing standard queries for the graph store."""
+
+    def __init__(
+        self,
+        dataset: Dataset,
+        default_named_graph: URIRef | None = None,
+    ) -> None:
+        self.select = SelectQueries(dataset, default_named_graph)
+        self.update = UpdateQueries(self.select, dataset, default_named_graph)

cognite_neat-0.119.2/cognite/neat/_graph/queries/_base.py → cognite_neat-0.119.3/cognite/neat/_graph/queries/_select.py

@@ -3,39 +3,30 @@ from collections import defaultdict
 from collections.abc import Iterable
 from typing import Any, Literal, cast, overload
 
-from rdflib import RDF, XSD, Dataset, Graph, Namespace, URIRef
+from rdflib import RDF, XSD, Namespace, URIRef
 from rdflib import Literal as RdfLiteral
-from rdflib.graph import DATASET_DEFAULT_GRAPH_ID
 from rdflib.query import ResultRow
 
 from cognite.neat._constants import NEAT
 from cognite.neat._shared import InstanceType
-from cognite.neat._utils.rdf_ import remove_instance_ids_in_batch, remove_namespace_from_uri
+from cognite.neat._utils.rdf_ import remove_namespace_from_uri
 
+from ._base import BaseQuery
 
-class Queries:
-    """Helper class for storing standard queries for the graph store."""
 
-    def __init__(
-        self,
-        dataset: Dataset,
-        default_named_graph: URIRef | None = None,
-    ):
-        self.dataset = dataset
-        self.default_named_graph = default_named_graph or DATASET_DEFAULT_GRAPH_ID
-
-    def graph(self, named_graph: URIRef | None = None) -> Graph:
-        """Get named graph from the dataset to query over"""
-        return self.dataset.graph(named_graph or self.default_named_graph)
+class SelectQueries(BaseQuery):
+    """This class holds a set of SPARQL queries which are reading triples from the knowledge graph.
+    The read queries are executed against query endpoint, and typically start with SELECT statement.
+    """
 
     def summarize_instances(self, named_graph: URIRef | None = None) -> list[tuple[str, int]]:
         """Summarize instances in the graph store by class and count"""
         query_statement = """ SELECT ?class (COUNT(?instance) AS ?instanceCount)
-                             WHERE {
-                             ?instance a ?class .
-                             }
-                             GROUP BY ?class
-                             ORDER BY DESC(?instanceCount) """
+                                 WHERE {
+                                 ?instance a ?class .
+                                 }
+                                 GROUP BY ?class
+                                 ORDER BY DESC(?instanceCount) """
 
         return [  # type: ignore[misc]
             (
@@ -48,7 +39,7 @@ class Queries:
     def types(self, named_graph: URIRef | None = None) -> dict[URIRef, str]:
         """Types and their short form in the graph"""
         query = """SELECT DISTINCT ?type
-                   WHERE {?s a ?type .}"""
+                       WHERE {?s a ?type .}"""
 
         return {  # type: ignore[misc, index, arg-type]
             cast(URIRef, type_): remove_namespace_from_uri(cast(URIRef, type_))
@@ -67,7 +58,7 @@ class Queries:
 
         """
         query = """SELECT DISTINCT ?property
-               WHERE {?s ?property ?o . FILTER(?property != rdf:type)}"""
+                   WHERE {?s ?property ?o . FILTER(?property != rdf:type)}"""
         return {  # type: ignore[misc, index, arg-type]
             cast(URIRef, type_): remove_namespace_from_uri(cast(URIRef, type_))
             for (type_,) in list(self.graph(named_graph).query(query))
@@ -81,7 +72,7 @@ class Queries:
 
         """
         query = """SELECT DISTINCT ?type ?property
-               WHERE {?s a ?type . ?s ?property ?o . FILTER(?property != rdf:type)}"""
+                   WHERE {?s a ?type . ?s ?property ?o . FILTER(?property != rdf:type)}"""
         properties_by_type: dict[URIRef, dict[URIRef, str]] = defaultdict(dict)
         for type_, property_ in cast(ResultRow, list(self.graph(named_graph).query(query))):
             properties_by_type[type_][property_] = remove_namespace_from_uri(property_)  # type: ignore[index]
@@ -305,21 +296,21 @@ class Queries:
         named_graph: URIRef | None = None,
     ) -> Iterable[tuple[URIRef, URIRef, list[URIRef]]]:
         query = """SELECT ?sourceType ?property
-                          (GROUP_CONCAT(DISTINCT STR(?valueType); SEPARATOR=",") AS ?valueTypes)
+                              (GROUP_CONCAT(DISTINCT STR(?valueType); SEPARATOR=",") AS ?valueTypes)
 
-                   WHERE {{
-                       ?s ?property ?o .
-                       ?s a ?sourceType .
-                       OPTIONAL {{ ?o a ?type }}
+                       WHERE {{
+                           ?s ?property ?o .
+                           ?s a ?sourceType .
+                           OPTIONAL {{ ?o a ?type }}
 
-                       # Key part to determine value type: either object, data or unknown
-                       BIND(   IF(isLiteral(?o),DATATYPE(?o),
-                               IF(BOUND(?type), ?type,
-                                               <{unknownType}>)) AS ?valueType)
-                   }}
+                           # Key part to determine value type: either object, data or unknown
+                           BIND(   IF(isLiteral(?o),DATATYPE(?o),
+                                   IF(BOUND(?type), ?type,
+                                                   <{unknownType}>)) AS ?valueType)
+                       }}
 
-                   GROUP BY ?sourceType ?property
-                   HAVING (COUNT(DISTINCT ?valueType) > 1)"""
+                       GROUP BY ?sourceType ?property
+                       HAVING (COUNT(DISTINCT ?valueType) > 1)"""
 
         for (
             source_type,
@@ -331,27 +322,6 @@ class Queries:
         ):
             yield cast(URIRef, source_type), cast(URIRef, property_), [URIRef(uri) for uri in value_types.split(",")]
 
-    def drop_types(
-        self,
-        type_: list[URIRef],
-        named_graph: URIRef | None = None,
-    ) -> dict[URIRef, int]:
-        """Drop types from the graph store
-
-        Args:
-            type_: List of types to drop
-            named_graph: Named graph to query over, default None (default graph
-
-        Returns:
-            Dictionary of dropped types
-        """
-        dropped_types: dict[URIRef, int] = {}
-        for t in type_:
-            instance_ids = list(self.list_instances_ids(t))
-            dropped_types[t] = len(instance_ids)
-            remove_instance_ids_in_batch(self.graph(named_graph), instance_ids)
-        return dropped_types
-
     def multi_type_instances(self, named_graph: URIRef | None = None) -> dict[str, list[str]]:
         """Find instances with multiple types
 
@@ -361,13 +331,13 @@ class Queries:
         """
 
         query = """
-        SELECT ?instance (GROUP_CONCAT(str(?type); SEPARATOR=",") AS ?types)
-        WHERE {
-            ?instance a ?type .
-        }
-        GROUP BY ?instance
-        HAVING (COUNT(?type) > 1)
-        """
+            SELECT ?instance (GROUP_CONCAT(str(?type); SEPARATOR=",") AS ?types)
+            WHERE {
+                ?instance a ?type .
+            }
+            GROUP BY ?instance
+            HAVING (COUNT(?type) > 1)
+            """
 
         result = {}
         for instance, types in self.graph(named_graph).query(query):  # type: ignore
@@ -383,14 +353,14 @@ class Queries:
         self, remove_namespace: bool = True, named_graph: URIRef | None = None
     ) -> list[dict[str, Any]]:
         query = """
-        SELECT ?type (COUNT(DISTINCT ?instance) AS ?instanceCount) (COUNT(DISTINCT ?property) AS ?propertyCount)
-                WHERE {
-                  ?instance a ?type .
-                  ?instance ?property ?value .
-                  FILTER(?property != rdf:type)
-                }
-                GROUP BY ?type
-                ORDER BY DESC(?instanceCount)"""
+            SELECT ?type (COUNT(DISTINCT ?instance) AS ?instanceCount) (COUNT(DISTINCT ?property) AS ?propertyCount)
+                    WHERE {
+                      ?instance a ?type .
+                      ?instance ?property ?value .
+                      FILTER(?property != rdf:type)
+                    }
+                    GROUP BY ?type
+                    ORDER BY DESC(?instanceCount)"""
         return [
             {
                 "type": urllib.parse.unquote(remove_namespace_from_uri(type_)) if remove_namespace else type_,
@@ -410,13 +380,13 @@ class Queries:
             for entry in self.types_with_instance_and_property_count(remove_namespace=False, named_graph=named_graph)
         }
         query = """SELECT ?type ?property (COUNT(DISTINCT ?instance) AS ?instanceCount)
-WHERE {
-  ?instance a ?type .
-  ?instance ?property ?value .
-  FILTER(?property != rdf:type)
-}
-GROUP BY ?type ?property
-ORDER BY ASC(?type) ASC(?property)"""
+    WHERE {
+      ?instance a ?type .
+      ?instance ?property ?value .
+      FILTER(?property != rdf:type)
+    }
+    GROUP BY ?type ?property
+    ORDER BY ASC(?type) ASC(?property)"""
         return [
             {
                 "type": urllib.parse.unquote(remove_namespace_from_uri(type_)) if remove_namespace else type_,
@@ -441,11 +411,11 @@ ORDER BY ASC(?type) ASC(?property)"""
         self, type: URIRef, remove_namespace: bool = True, named_graph: URIRef | None = None
     ) -> dict[str, set[str]] | dict[URIRef, set[URIRef]]:
         query = """SELECT DISTINCT ?instance ?property
-WHERE {{
-    ?instance a <{type}> .
-    ?instance ?property ?value .
-    FILTER(?property != rdf:type)
-}}"""
+    WHERE {{
+        ?instance a <{type}> .
+        ?instance ?property ?value .
+        FILTER(?property != rdf:type)
+    }}"""
         result = defaultdict(set)
         for instance, property_ in cast(Iterable[ResultRow], self.graph(named_graph).query(query.format(type=type))):
             instance_str = urllib.parse.unquote(remove_namespace_from_uri(instance)) if remove_namespace else instance
@@ -458,7 +428,7 @@ WHERE {{
     ) -> Iterable[tuple[URIRef, str]]:
         """Returns instance ids by space"""
         query = f"""SELECT DISTINCT ?instance ?space
-                   WHERE {{?instance <{space_property}> ?space}}"""
+                       WHERE {{?instance <{space_property}> ?space}}"""
 
         for result in cast(Iterable[ResultRow], self.graph(named_graph).query(query)):
             instance_id, space = cast(tuple[URIRef, URIRef | RdfLiteral], result)

cognite_neat-0.119.3/cognite/neat/_graph/queries/_update.py

@@ -0,0 +1,37 @@
+from rdflib import Dataset, URIRef
+
+from cognite.neat._utils.rdf_ import remove_instance_ids_in_batch
+
+from ._base import BaseQuery
+from ._select import SelectQueries
+
+
+class UpdateQueries(BaseQuery):
+    """This class holds a set of SPARQL queries which are updating triples in the knowledge graph.
+    The update queries are executed against update endpoint, and typically start with UPDATE statement
+    """
+
+    def __init__(self, read: SelectQueries, dataset: Dataset, default_named_graph: URIRef | None = None) -> None:
+        super().__init__(dataset, default_named_graph)
+        self._read = read
+
+    def drop_types(
+        self,
+        type_: list[URIRef],
+        named_graph: URIRef | None = None,
+    ) -> dict[URIRef, int]:
+        """Drop types from the graph store
+
+        Args:
+            type_: List of types to drop
+            named_graph: Named graph to query over, default None (default graph
+
+        Returns:
+            Dictionary of dropped types
+        """
+        dropped_types: dict[URIRef, int] = {}
+        for t in type_:
+            instance_ids = list(self._read.list_instances_ids(t))
+            dropped_types[t] = len(instance_ids)
+            remove_instance_ids_in_batch(self.graph(named_graph), instance_ids)
+        return dropped_types