cognee 0.5.0.dev0__py3-none-any.whl → 0.5.0.dev1__py3-none-any.whl

cognee/api/client.py

@@ -21,7 +21,7 @@ from cognee.api.v1.notebooks.routers import get_notebooks_router
 from cognee.api.v1.permissions.routers import get_permissions_router
 from cognee.api.v1.settings.routers import get_settings_router
 from cognee.api.v1.datasets.routers import get_datasets_router
-from cognee.api.v1.cognify.routers import get_code_pipeline_router, get_cognify_router
+from cognee.api.v1.cognify.routers import get_cognify_router
 from cognee.api.v1.search.routers import get_search_router
 from cognee.api.v1.ontologies.routers.get_ontology_router import get_ontology_router
 from cognee.api.v1.memify.routers import get_memify_router
@@ -278,10 +278,6 @@ app.include_router(get_responses_router(), prefix="/api/v1/responses", tags=["re
 
 app.include_router(get_sync_router(), prefix="/api/v1/sync", tags=["sync"])
 
-codegraph_routes = get_code_pipeline_router()
-if codegraph_routes:
-    app.include_router(codegraph_routes, prefix="/api/v1/code-pipeline", tags=["code-pipeline"])
-
 app.include_router(
     get_users_router(),
     prefix="/api/v1/users",

cognee/api/v1/add/add.py

@@ -155,7 +155,7 @@ async def add(
         - LLM_API_KEY: API key for your LLM provider (OpenAI, Anthropic, etc.)
 
         Optional:
-        - LLM_PROVIDER: "openai" (default), "anthropic", "gemini", "ollama", "mistral"
+        - LLM_PROVIDER: "openai" (default), "anthropic", "gemini", "ollama", "mistral", "bedrock"
         - LLM_MODEL: Model name (default: "gpt-5-mini")
         - DEFAULT_USER_EMAIL: Custom default user email
         - DEFAULT_USER_PASSWORD: Custom default user password
@@ -205,6 +205,7 @@ async def add(
         pipeline_name="add_pipeline",
         vector_db_config=vector_db_config,
         graph_db_config=graph_db_config,
+        use_pipeline_cache=True,
         incremental_loading=incremental_loading,
         data_per_batch=data_per_batch,
     ):

cognee/api/v1/cognify/cognify.py

@@ -3,6 +3,7 @@ from pydantic import BaseModel
 from typing import Union, Optional
 from uuid import UUID
 
+from cognee.modules.cognify.config import get_cognify_config
 from cognee.modules.ontology.ontology_env_config import get_ontology_env_config
 from cognee.shared.logging_utils import get_logger
 from cognee.shared.data_models import KnowledgeGraph
@@ -19,7 +20,6 @@ from cognee.modules.ontology.get_default_ontology_resolver import (
 from cognee.modules.users.models import User
 
 from cognee.tasks.documents import (
-    check_permissions_on_dataset,
     classify_documents,
     extract_chunks_from_documents,
 )
@@ -53,6 +53,7 @@ async def cognify(
     custom_prompt: Optional[str] = None,
     temporal_cognify: bool = False,
     data_per_batch: int = 20,
+    **kwargs,
 ):
     """
     Transform ingested data into a structured knowledge graph.
@@ -78,12 +79,11 @@ async def cognify(
 
     Processing Pipeline:
         1. **Document Classification**: Identifies document types and structures
-        2. **Permission Validation**: Ensures user has processing rights
-        3. **Text Chunking**: Breaks content into semantically meaningful segments
-        4. **Entity Extraction**: Identifies key concepts, people, places, organizations
-        5. **Relationship Detection**: Discovers connections between entities
-        6. **Graph Construction**: Builds semantic knowledge graph with embeddings
-        7. **Content Summarization**: Creates hierarchical summaries for navigation
+        2. **Text Chunking**: Breaks content into semantically meaningful segments
+        3. **Entity Extraction**: Identifies key concepts, people, places, organizations
+        4. **Relationship Detection**: Discovers connections between entities
+        5. **Graph Construction**: Builds semantic knowledge graph with embeddings
+        6. **Content Summarization**: Creates hierarchical summaries for navigation
 
     Graph Model Customization:
         The `graph_model` parameter allows custom knowledge structures:
@@ -224,6 +224,7 @@ async def cognify(
             config=config,
             custom_prompt=custom_prompt,
             chunks_per_batch=chunks_per_batch,
+            **kwargs,
         )
 
     # By calling get pipeline executor we get a function that will have the run_pipeline run in the background or a function that we will need to wait for
@@ -238,6 +239,7 @@ async def cognify(
         vector_db_config=vector_db_config,
         graph_db_config=graph_db_config,
         incremental_loading=incremental_loading,
+        use_pipeline_cache=True,
         pipeline_name="cognify_pipeline",
         data_per_batch=data_per_batch,
     )
@@ -251,6 +253,7 @@ async def get_default_tasks(  # TODO: Find out a better way to do this (Boris's
     config: Config = None,
     custom_prompt: Optional[str] = None,
     chunks_per_batch: int = 100,
+    **kwargs,
 ) -> list[Task]:
     if config is None:
         ontology_config = get_ontology_env_config()
@@ -272,9 +275,11 @@ async def get_default_tasks(  # TODO: Find out a better way to do this (Boris's
     if chunks_per_batch is None:
         chunks_per_batch = 100
 
+    cognify_config = get_cognify_config()
+    embed_triplets = cognify_config.triplet_embedding
+
     default_tasks = [
         Task(classify_documents),
-        Task(check_permissions_on_dataset, user=user, permissions=["write"]),
         Task(
             extract_chunks_from_documents,
             max_chunk_size=chunk_size or get_max_chunk_tokens(),
@@ -286,12 +291,17 @@ async def get_default_tasks(  # TODO: Find out a better way to do this (Boris's
             config=config,
             custom_prompt=custom_prompt,
             task_config={"batch_size": chunks_per_batch},
+            **kwargs,
         ),  # Generate knowledge graphs from the document chunks.
         Task(
             summarize_text,
             task_config={"batch_size": chunks_per_batch},
         ),
-        Task(add_data_points, task_config={"batch_size": chunks_per_batch}),
+        Task(
+            add_data_points,
+            embed_triplets=embed_triplets,
+            task_config={"batch_size": chunks_per_batch},
+        ),
     ]
 
     return default_tasks
@@ -305,14 +315,13 @@ async def get_temporal_tasks(
 
     The pipeline includes:
     1. Document classification.
-    2. Dataset permission checks (requires "write" access).
-    3. Document chunking with a specified or default chunk size.
-    4. Event and timestamp extraction from chunks.
-    5. Knowledge graph extraction from events.
-    6. Batched insertion of data points.
+    2. Document chunking with a specified or default chunk size.
+    3. Event and timestamp extraction from chunks.
+    4. Knowledge graph extraction from events.
+    5. Batched insertion of data points.
 
     Args:
-        user (User, optional): The user requesting task execution, used for permission checks.
+        user (User, optional): The user requesting task execution.
         chunker (Callable, optional): A text chunking function/class to split documents. Defaults to TextChunker.
         chunk_size (int, optional): Maximum token size per chunk. If not provided, uses system default.
         chunks_per_batch (int, optional): Number of chunks to process in a single batch in Cognify
@@ -325,7 +334,6 @@ async def get_temporal_tasks(
 
     temporal_tasks = [
         Task(classify_documents),
-        Task(check_permissions_on_dataset, user=user, permissions=["write"]),
         Task(
             extract_chunks_from_documents,
             max_chunk_size=chunk_size or get_max_chunk_tokens(),

cognee/api/v1/cognify/routers/__init__.py

@@ -1,2 +1 @@
 from .get_cognify_router import get_cognify_router
-from .get_code_pipeline_router import get_code_pipeline_router

cognee/api/v1/cognify/routers/get_cognify_router.py

@@ -42,7 +42,9 @@ class CognifyPayloadDTO(InDTO):
         default="", description="Custom prompt for entity extraction and graph generation"
     )
     ontology_key: Optional[List[str]] = Field(
-        default=None, description="Reference to one or more previously uploaded ontologies"
+        default=None,
+        examples=[[]],
+        description="Reference to one or more previously uploaded ontologies",
     )
 
 

cognee/api/v1/datasets/routers/get_datasets_router.py

@@ -208,14 +208,14 @@ def get_datasets_router() -> APIRouter:
             },
         )
 
-        from cognee.modules.data.methods import get_dataset, delete_dataset
+        from cognee.modules.data.methods import delete_dataset
 
-        dataset = await get_dataset(user.id, dataset_id)
+        dataset = await get_authorized_existing_datasets([dataset_id], "delete", user)
 
         if dataset is None:
             raise DatasetNotFoundError(message=f"Dataset ({str(dataset_id)}) not found.")
 
-        await delete_dataset(dataset)
+        await delete_dataset(dataset[0])
 
     @router.delete(
         "/{dataset_id}/data/{data_id}",

cognee/api/v1/ontologies/ontologies.py

@@ -5,6 +5,7 @@ from pathlib import Path
 from datetime import datetime, timezone
 from typing import Optional, List
 from dataclasses import dataclass
+from fastapi import UploadFile
 
 
 @dataclass
@@ -45,8 +46,10 @@ class OntologyService:
             json.dump(metadata, f, indent=2)
 
     async def upload_ontology(
-        self, ontology_key: str, file, user, description: Optional[str] = None
+        self, ontology_key: str, file: UploadFile, user, description: Optional[str] = None
     ) -> OntologyMetadata:
+        if not file.filename:
+            raise ValueError("File must have a filename")
         if not file.filename.lower().endswith(".owl"):
             raise ValueError("File must be in .owl format")
 
@@ -57,8 +60,6 @@ class OntologyService:
             raise ValueError(f"Ontology key '{ontology_key}' already exists")
 
         content = await file.read()
-        if len(content) > 10 * 1024 * 1024:
-            raise ValueError("File size exceeds 10MB limit")
 
         file_path = user_dir / f"{ontology_key}.owl"
         with open(file_path, "wb") as f:
@@ -82,7 +83,11 @@ class OntologyService:
         )
 
     async def upload_ontologies(
-        self, ontology_key: List[str], files: List, user, descriptions: Optional[List[str]] = None
+        self,
+        ontology_key: List[str],
+        files: List[UploadFile],
+        user,
+        descriptions: Optional[List[str]] = None,
     ) -> List[OntologyMetadata]:
         """
         Upload ontology files with their respective keys.
@@ -105,47 +110,17 @@ class OntologyService:
         if len(set(ontology_key)) != len(ontology_key):
             raise ValueError("Duplicate ontology keys not allowed")
 
-        if descriptions and len(descriptions) != len(files):
-            raise ValueError("Number of descriptions must match number of files")
-
         results = []
-        user_dir = self._get_user_dir(str(user.id))
-        metadata = self._load_metadata(user_dir)
 
         for i, (key, file) in enumerate(zip(ontology_key, files)):
-            if key in metadata:
-                raise ValueError(f"Ontology key '{key}' already exists")
-
-            if not file.filename.lower().endswith(".owl"):
-                raise ValueError(f"File '{file.filename}' must be in .owl format")
-
-            content = await file.read()
-            if len(content) > 10 * 1024 * 1024:
-                raise ValueError(f"File '{file.filename}' exceeds 10MB limit")
-
-            file_path = user_dir / f"{key}.owl"
-            with open(file_path, "wb") as f:
-                f.write(content)
-
-            ontology_metadata = {
-                "filename": file.filename,
-                "size_bytes": len(content),
-                "uploaded_at": datetime.now(timezone.utc).isoformat(),
-                "description": descriptions[i] if descriptions else None,
-            }
-            metadata[key] = ontology_metadata
-
             results.append(
-                OntologyMetadata(
+                await self.upload_ontology(
                     ontology_key=key,
-                    filename=file.filename,
-                    size_bytes=len(content),
-                    uploaded_at=ontology_metadata["uploaded_at"],
+                    file=file,
+                    user=user,
                     description=descriptions[i] if descriptions else None,
                 )
             )
-
-        self._save_metadata(user_dir, metadata)
         return results
 
     def get_ontology_contents(self, ontology_key: List[str], user) -> List[str]:

cognee/api/v1/ontologies/routers/get_ontology_router.py

@@ -1,4 +1,4 @@
-from fastapi import APIRouter, File, Form, UploadFile, Depends, HTTPException
+from fastapi import APIRouter, File, Form, UploadFile, Depends, Request
 from fastapi.responses import JSONResponse
 from typing import Optional, List
 
@@ -15,28 +15,25 @@ def get_ontology_router() -> APIRouter:
 
     @router.post("", response_model=dict)
     async def upload_ontology(
+        request: Request,
         ontology_key: str = Form(...),
-        ontology_file: List[UploadFile] = File(...),
-        descriptions: Optional[str] = Form(None),
+        ontology_file: UploadFile = File(...),
+        description: Optional[str] = Form(None),
         user: User = Depends(get_authenticated_user),
     ):
         """
-        Upload ontology files with their respective keys for later use in cognify operations.
-
-        Supports both single and multiple file uploads:
-        - Single file: ontology_key=["key"], ontology_file=[file]
-        - Multiple files: ontology_key=["key1", "key2"], ontology_file=[file1, file2]
+        Upload a single ontology file for later use in cognify operations.
 
         ## Request Parameters
-        - **ontology_key** (str): JSON array string of user-defined identifiers for the ontologies
-        - **ontology_file** (List[UploadFile]): OWL format ontology files
-        - **descriptions** (Optional[str]): JSON array string of optional descriptions
+        - **ontology_key** (str): User-defined identifier for the ontology.
+        - **ontology_file** (UploadFile): Single OWL format ontology file
+        - **description** (Optional[str]): Optional description for the ontology.
 
         ## Response
-        Returns metadata about uploaded ontologies including keys, filenames, sizes, and upload timestamps.
+        Returns metadata about the uploaded ontology including key, filename, size, and upload timestamp.
 
         ## Error Codes
-        - **400 Bad Request**: Invalid file format, duplicate keys, array length mismatches, file size exceeded
+        - **400 Bad Request**: Invalid file format, duplicate key, multiple files uploaded
         - **500 Internal Server Error**: File system or processing errors
         """
         send_telemetry(
@@ -49,16 +46,22 @@ def get_ontology_router() -> APIRouter:
         )
 
         try:
-            import json
-
-            ontology_keys = json.loads(ontology_key)
-            description_list = json.loads(descriptions) if descriptions else None
-
-            if not isinstance(ontology_keys, list):
-                raise ValueError("ontology_key must be a JSON array")
-
-            results = await ontology_service.upload_ontologies(
-                ontology_keys, ontology_file, user, description_list
+            # Enforce: exactly one uploaded file for "ontology_file"
+            form = await request.form()
+            uploaded_files = form.getlist("ontology_file")
+            if len(uploaded_files) != 1:
+                raise ValueError("Only one ontology_file is allowed")
+
+            if ontology_key.strip().startswith(("[", "{")):
+                raise ValueError("ontology_key must be a string")
+            if description is not None and description.strip().startswith(("[", "{")):
+                raise ValueError("description must be a string")
+
+            result = await ontology_service.upload_ontology(
+                ontology_key=ontology_key,
+                file=ontology_file,
+                user=user,
+                description=description,
             )
 
             return {
@@ -70,10 +73,9 @@ def get_ontology_router() -> APIRouter:
                         "uploaded_at": result.uploaded_at,
                         "description": result.description,
                     }
-                    for result in results
                 ]
             }
-        except (json.JSONDecodeError, ValueError) as e:
+        except ValueError as e:
             return JSONResponse(status_code=400, content={"error": str(e)})
         except Exception as e:
             return JSONResponse(status_code=500, content={"error": str(e)})

cognee/api/v1/search/search.py

@@ -31,6 +31,8 @@ async def search(
     only_context: bool = False,
     use_combined_context: bool = False,
     session_id: Optional[str] = None,
+    wide_search_top_k: Optional[int] = 100,
+    triplet_distance_penalty: Optional[float] = 3.5,
 ) -> Union[List[SearchResult], CombinedSearchResult]:
     """
     Search and query the knowledge graph for insights, information, and connections.
@@ -200,6 +202,8 @@ async def search(
         only_context=only_context,
         use_combined_context=use_combined_context,
         session_id=session_id,
+        wide_search_top_k=wide_search_top_k,
+        triplet_distance_penalty=triplet_distance_penalty,
     )
 
     return filtered_search_results