add merge

.github/actions/cognee_setup/action.yml

@@ -41,4 +41,4 @@ runs:
             EXTRA_ARGS="$EXTRA_ARGS --extra $extra"
           done
         fi
-        uv sync --extra api --extra docs --extra evals --extra codegraph --extra ollama --extra dev --extra neo4j $EXTRA_ARGS
+        uv sync --extra api --extra docs --extra evals --extra codegraph --extra ollama --extra dev --extra neo4j --extra redis $EXTRA_ARGS

.github/workflows/e2e_tests.yml

@@ -1,4 +1,6 @@
 name: Reusable Integration Tests
+permissions:
+  contents: read
 
 on:
   workflow_call:
@@ -264,3 +266,68 @@ jobs:
           EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
         run: uv run python ./cognee/tests/test_edge_ingestion.py
+
+
+
+  run_concurrent_subprocess_access_test:
+    name: Concurrent Subprocess access test
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        shell: bash
+    services:
+      postgres:
+        image: pgvector/pgvector:pg17
+        env:
+          POSTGRES_USER: cognee
+          POSTGRES_PASSWORD: cognee
+          POSTGRES_DB: cognee_db
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+        ports:
+          - 5432:5432
+
+      redis:
+        image: redis:7
+        ports:
+          - 6379:6379
+        options: >-
+          --health-cmd "redis-cli ping"
+          --health-interval 5s
+          --health-timeout 3s
+          --health-retries 5
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Cognee Setup
+        uses: ./.github/actions/cognee_setup
+        with:
+          python-version: '3.11.x'
+          extra-dependencies: "postgres redis"
+
+      - name: Run Concurrent subprocess access test (Kuzu/Lancedb/Postgres)
+        env:
+          ENV: dev
+          LLM_MODEL: ${{ secrets.LLM_MODEL }}
+          LLM_ENDPOINT: ${{ secrets.LLM_ENDPOINT }}
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_API_VERSION: ${{ secrets.LLM_API_VERSION }}
+          EMBEDDING_MODEL: ${{ secrets.EMBEDDING_MODEL }}
+          EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
+          EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
+          EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
+          GRAPH_DATABASE_PROVIDER: 'kuzu'
+          CACHING: true
+          SHARED_KUZU_LOCK: true
+          DB_PROVIDER: 'postgres'
+          DB_NAME: 'cognee_db'
+          DB_HOST: '127.0.0.1'
+          DB_PORT: 5432
+          DB_USERNAME: cognee
+          DB_PASSWORD: cognee
+        run: uv run python ./cognee/tests/test_concurrent_subprocess_access.py

.github/workflows/examples_tests.yml

@@ -110,6 +110,81 @@ jobs:
           EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
         run: uv run python ./examples/python/dynamic_steps_example.py
 
+  test-temporal-example:
+    name: Run Temporal Tests
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Check out repository
+        uses: actions/checkout@v4
+
+      - name: Cognee Setup
+        uses: ./.github/actions/cognee_setup
+        with:
+          python-version: '3.11.x'
+
+      - name: Run Temporal Example
+        env:
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          LLM_MODEL: ${{ secrets.LLM_MODEL }}
+          LLM_ENDPOINT: ${{ secrets.LLM_ENDPOINT }}
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_API_VERSION: ${{ secrets.LLM_API_VERSION }}
+          EMBEDDING_MODEL: ${{ secrets.EMBEDDING_MODEL }}
+          EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
+          EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
+          EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
+        run: uv run python ./examples/python/temporal_example.py
+
+  test-ontology-example:
+    name: Run Ontology Tests
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Check out repository
+        uses: actions/checkout@v4
+
+      - name: Cognee Setup
+        uses: ./.github/actions/cognee_setup
+        with:
+          python-version: '3.11.x'
+
+      - name: Run Ontology Demo Example
+        env:
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          LLM_MODEL: ${{ secrets.LLM_MODEL }}
+          LLM_ENDPOINT: ${{ secrets.LLM_ENDPOINT }}
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_API_VERSION: ${{ secrets.LLM_API_VERSION }}
+          EMBEDDING_MODEL: ${{ secrets.EMBEDDING_MODEL }}
+          EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
+          EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
+          EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
+        run: uv run python ./examples/python/ontology_demo_example.py
+
+  test-agentic-reasoning:
+    name: Run Agentic Reasoning Tests
+    runs-on: ubuntu-22.04
+    steps:
+      - name: Check out repository
+        uses: actions/checkout@v4
+
+      - name: Cognee Setup
+        uses: ./.github/actions/cognee_setup
+        with:
+          python-version: '3.11.x'
+
+      - name: Run Agentic Reasoning Example
+        env:
+          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
+          LLM_MODEL: ${{ secrets.LLM_MODEL }}
+          LLM_ENDPOINT: ${{ secrets.LLM_ENDPOINT }}
+          LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
+          LLM_API_VERSION: ${{ secrets.LLM_API_VERSION }}
+          EMBEDDING_MODEL: ${{ secrets.EMBEDDING_MODEL }}
+          EMBEDDING_ENDPOINT: ${{ secrets.EMBEDDING_ENDPOINT }}
+          EMBEDDING_API_KEY: ${{ secrets.EMBEDDING_API_KEY }}
+          EMBEDDING_API_VERSION: ${{ secrets.EMBEDDING_API_VERSION }}
+        run: uv run python ./examples/python/agentic_reasoning_procurement_example.py
+
   test-memify:
     name: Run Memify Example
     runs-on: ubuntu-22.04

.github/workflows/test_different_operating_systems.yml

@@ -9,7 +9,7 @@ on:
       python-versions:
         required: false
         type: string
-        default: '["3.10.x", "3.11.x", "3.12.x"]'
+        default: '["3.10.x", "3.12.x", "3.13.x"]'
     secrets:
       LLM_PROVIDER:
         required: true
@@ -193,6 +193,13 @@ jobs:
         with:
           python-version: ${{ matrix.python-version }}
 
+      - name: Path setup
+        if: ${{ matrix.os }} == 'windows-latest'
+        shell: bash
+        run: |
+          PATH=$(printf '%s' "$PATH" | tr ':' $'\n' | grep -vi '/git/usr/bin' | paste -sd: -)
+          export PATH
+
       - name: Run Soft Deletion Tests
         env:
           ENV: 'dev'

.github/workflows/test_suites.yml

@@ -85,7 +85,7 @@ jobs:
     needs: [basic-tests, e2e-tests]
     uses: ./.github/workflows/test_different_operating_systems.yml
     with:
-      python-versions: '["3.10.x", "3.11.x", "3.12.x"]'
+      python-versions: '["3.10.x", "3.11.x", "3.12.x", "3.13.x"]'
     secrets: inherit
 
   # Matrix-based vector database tests

cognee/api/v1/add/add.py

@@ -41,6 +41,7 @@ async def add(
     extraction_rules: Optional[Dict[str, Any]] = None,
     tavily_config: Optional[BaseModel] = None,
     soup_crawler_config: Optional[BaseModel] = None,
+    data_per_batch: Optional[int] = 20,
 ):
     """
     Add data to Cognee for knowledge graph processing.
@@ -235,6 +236,7 @@ async def add(
         vector_db_config=vector_db_config,
         graph_db_config=graph_db_config,
         incremental_loading=incremental_loading,
+        data_per_batch=data_per_batch,
     ):
         pipeline_run_info = run_info
 

cognee/api/v1/cognify/cognify.py

@@ -52,6 +52,7 @@ async def cognify(
     incremental_loading: bool = True,
     custom_prompt: Optional[str] = None,
     temporal_cognify: bool = False,
+    data_per_batch: int = 20,
 ):
     """
     Transform ingested data into a structured knowledge graph.
@@ -238,6 +239,7 @@ async def cognify(
         graph_db_config=graph_db_config,
         incremental_loading=incremental_loading,
         pipeline_name="cognify_pipeline",
+        data_per_batch=data_per_batch,
     )
 
 

cognee/api/v1/search/search.py

@@ -1,6 +1,7 @@
 from uuid import UUID
 from typing import Union, Optional, List, Type
 
+from cognee.infrastructure.databases.graph import get_graph_engine
 from cognee.modules.engine.models.node_set import NodeSet
 from cognee.modules.users.models import User
 from cognee.modules.search.types import SearchResult, SearchType, CombinedSearchResult
@@ -8,6 +9,9 @@ from cognee.modules.users.methods import get_default_user
 from cognee.modules.search.methods import search as search_function
 from cognee.modules.data.methods import get_authorized_existing_datasets
 from cognee.modules.data.exceptions import DatasetNotFoundError
+from cognee.shared.logging_utils import get_logger
+
+logger = get_logger()
 
 
 async def search(
@@ -175,6 +179,13 @@ async def search(
         if not datasets:
             raise DatasetNotFoundError(message="No datasets found.")
 
+    graph_engine = await get_graph_engine()
+    is_empty = await graph_engine.is_empty()
+
+    if is_empty:
+        logger.warning("Search attempt on an empty knowledge graph")
+        return []
+
     filtered_search_results = await search_function(
         query_text=query_text,
         query_type=query_type,

cognee/infrastructure/databases/cache/__init__.py

@@ -0,0 +1,2 @@
+from .get_cache_engine import get_cache_engine
+from .config import get_cache_config

cognee/infrastructure/databases/cache/cache_db_interface.py

@@ -0,0 +1,42 @@
+from abc import ABC, abstractmethod
+from contextlib import contextmanager
+
+
+class CacheDBInterface(ABC):
+    """
+    Abstract base class for distributed cache coordination systems (e.g., Redis, Memcached).
+    Provides a common interface for lock acquisition, release, and context-managed locking.
+    """
+
+    def __init__(self, host: str, port: int, lock_key: str):
+        self.host = host
+        self.port = port
+        self.lock_key = lock_key
+        self.lock = None
+
+    @abstractmethod
+    def acquire_lock(self):
+        """
+        Acquire a lock on the given key.
+        Must be implemented by subclasses.
+        """
+        pass
+
+    @abstractmethod
+    def release_lock(self):
+        """
+        Release the lock if it is held.
+        Must be implemented by subclasses.
+        """
+        pass
+
+    @contextmanager
+    def hold_lock(self):
+        """
+        Context manager for safely acquiring and releasing the lock.
+        """
+        self.acquire()
+        try:
+            yield
+        finally:
+            self.release()

cognee/infrastructure/databases/cache/config.py

@@ -0,0 +1,39 @@
+from pydantic_settings import BaseSettings, SettingsConfigDict
+from functools import lru_cache
+
+
+class CacheConfig(BaseSettings):
+    """
+    Configuration for distributed cache systems (e.g., Redis), used for locking or coordination.
+
+    Attributes:
+    - shared_kuzu_lock: Shared kuzu lock logic on/off.
+    - cache_host: Hostname of the cache service.
+    - cache_port: Port number for the cache service.
+    - agentic_lock_expire: Automatic lock expiration time (in seconds).
+    - agentic_lock_timeout: Maximum time (in seconds) to wait for the lock release.
+    """
+
+    caching: bool = False
+    shared_kuzu_lock: bool = False
+    cache_host: str = "localhost"
+    cache_port: int = 6379
+    agentic_lock_expire: int = 240
+    agentic_lock_timeout: int = 300
+
+    model_config = SettingsConfigDict(env_file=".env", extra="allow")
+
+    def to_dict(self) -> dict:
+        return {
+            "caching": self.caching,
+            "shared_kuzu_lock": self.shared_kuzu_lock,
+            "cache_host": self.cache_host,
+            "cache_port": self.cache_port,
+            "agentic_lock_expire": self.agentic_lock_expire,
+            "agentic_lock_timeout": self.agentic_lock_timeout,
+        }
+
+
+@lru_cache
+def get_cache_config():
+    return CacheConfig()

cognee/infrastructure/databases/cache/get_cache_engine.py

@@ -0,0 +1,59 @@
+"""Factory to get the appropriate cache coordination engine (e.g., Redis)."""
+
+from functools import lru_cache
+from cognee.infrastructure.databases.cache.config import get_cache_config
+
+from cognee.infrastructure.databases.cache.cache_db_interface import CacheDBInterface
+
+config = get_cache_config()
+
+
+@lru_cache
+def create_cache_engine(
+    cache_host: str,
+    cache_port: int,
+    lock_key: str,
+    agentic_lock_expire: int = 240,
+    agentic_lock_timeout: int = 300,
+):
+    """
+    Factory function to instantiate a cache coordination backend (currently Redis).
+
+    Parameters:
+    -----------
+    - cache_host: Hostname or IP of the cache server.
+    - cache_port: Port number to connect to.
+    - lock_key: Identifier used for the locking resource.
+    - agentic_lock_expire: Duration to hold the lock after acquisition.
+    - agentic_lock_timeout: Max time to wait for the lock before failing.
+
+    Returns:
+    --------
+    - CacheDBInterface: An instance of the appropriate cache adapter. :TODO: Now we support only Redis. later if we add more here we can split the logic
+    """
+    if config.caching:
+        from cognee.infrastructure.databases.cache.redis.RedisAdapter import RedisAdapter
+
+        return RedisAdapter(
+            host=cache_host,
+            port=cache_port,
+            lock_name=lock_key,
+            timeout=agentic_lock_expire,
+            blocking_timeout=agentic_lock_timeout,
+        )
+    else:
+        return None
+
+
+def get_cache_engine(lock_key: str) -> CacheDBInterface:
+    """
+    Returns a cache adapter instance using current context configuration.
+    """
+
+    return create_cache_engine(
+        cache_host=config.cache_host,
+        cache_port=config.cache_port,
+        lock_key=lock_key,
+        agentic_lock_expire=config.agentic_lock_expire,
+        agentic_lock_timeout=config.agentic_lock_timeout,
+    )

cognee/infrastructure/databases/cache/redis/RedisAdapter.py

@@ -0,0 +1,49 @@
+import redis
+from contextlib import contextmanager
+from cognee.infrastructure.databases.cache.cache_db_interface import CacheDBInterface
+
+
+class RedisAdapter(CacheDBInterface):
+    def __init__(self, host, port, lock_name, timeout=240, blocking_timeout=300):
+        super().__init__(host, port, lock_name)
+        self.redis = redis.Redis(host=host, port=port)
+        self.timeout = timeout
+        self.blocking_timeout = blocking_timeout
+
+    def acquire_lock(self):
+        """
+        Acquire the Redis lock manually. Raises if acquisition fails.
+        """
+        self.lock = self.redis.lock(
+            name=self.lock_key,
+            timeout=self.timeout,
+            blocking_timeout=self.blocking_timeout,
+        )
+
+        acquired = self.lock.acquire()
+        if not acquired:
+            raise RuntimeError(f"Could not acquire Redis lock: {self.lock_key}")
+
+        return self.lock
+
+    def release_lock(self):
+        """
+        Release the Redis lock manually, if held.
+        """
+        if self.lock:
+            try:
+                self.lock.release()
+                self.lock = None
+            except redis.exceptions.LockError:
+                pass
+
+    @contextmanager
+    def hold_lock(self):
+        """
+        Context manager for acquiring and releasing the Redis lock automatically.
+        """
+        self.acquire()
+        try:
+            yield
+        finally:
+            self.release()

cognee/infrastructure/databases/graph/get_graph_engine.py

@@ -162,5 +162,5 @@ def create_graph_engine(
 
     raise EnvironmentError(
         f"Unsupported graph database provider: {graph_database_provider}. "
-        f"Supported providers are: {', '.join(list(supported_databases.keys()) + ['neo4j', 'kuzu', 'kuzu-remote', 'memgraph', 'neptune', 'neptune_analytics'])}"
+        f"Supported providers are: {', '.join(list(supported_databases.keys()) + ['neo4j', 'kuzu', 'kuzu-remote', 'neptune', 'neptune_analytics'])}"
     )

cognee/infrastructure/databases/graph/graph_db_interface.py

@@ -159,6 +159,11 @@ class GraphDBInterface(ABC):
     - get_connections
     """
 
+    @abstractmethod
+    async def is_empty(self) -> bool:
+        logger.warning("is_empty() is not implemented")
+        return True
+
     @abstractmethod
     async def query(self, query: str, params: dict) -> List[Any]:
         """

cognee/infrastructure/databases/graph/kuzu/adapter.py

@@ -4,7 +4,7 @@ import os
 import json
 import asyncio
 import tempfile
-from uuid import UUID
+from uuid import UUID, uuid5, NAMESPACE_OID
 from kuzu import Connection
 from kuzu.database import Database
 from datetime import datetime, timezone
@@ -23,9 +23,14 @@ from cognee.infrastructure.engine import DataPoint
 from cognee.modules.storage.utils import JSONEncoder
 from cognee.modules.engine.utils.generate_timestamp_datapoint import date_to_int
 from cognee.tasks.temporal_graph.models import Timestamp
+from cognee.infrastructure.databases.cache.config import get_cache_config
 
 logger = get_logger()
 
+cache_config = get_cache_config()
+if cache_config.shared_kuzu_lock:
+    from cognee.infrastructure.databases.cache.get_cache_engine import get_cache_engine
+
 
 class KuzuAdapter(GraphDBInterface):
     """
@@ -39,12 +44,20 @@ class KuzuAdapter(GraphDBInterface):
 
     def __init__(self, db_path: str):
         """Initialize Kuzu database connection and schema."""
+        self.open_connections = 0
+        self._is_closed = False
         self.db_path = db_path  # Path for the database directory
         self.db: Optional[Database] = None
         self.connection: Optional[Connection] = None
+        if cache_config.shared_kuzu_lock:
+            self.redis_lock = get_cache_engine(
+                lock_key="kuzu-lock-" + str(uuid5(NAMESPACE_OID, db_path))
+            )
+        else:
             self.executor = ThreadPoolExecutor()
             self._initialize_connection()
         self.KUZU_ASYNC_LOCK = asyncio.Lock()
+        self._connection_change_lock = asyncio.Lock()
 
     def _initialize_connection(self) -> None:
         """Initialize the Kuzu database connection and schema."""
@@ -185,6 +198,15 @@ class KuzuAdapter(GraphDBInterface):
         except FileNotFoundError:
             logger.warning(f"Kuzu S3 storage file not found: {self.db_path}")
 
+    async def is_empty(self) -> bool:
+        query = """
+        MATCH (n)
+        RETURN true
+        LIMIT 1;
+        """
+        query_result = await self.query(query)
+        return len(query_result) == 0
+
     async def query(self, query: str, params: Optional[dict] = None) -> List[Tuple]:
         """
         Execute a Kuzu query asynchronously with automatic reconnection.
@@ -209,9 +231,13 @@ class KuzuAdapter(GraphDBInterface):
         params = params or {}
 
         def blocking_query():
+            lock_acquired = False
             try:
+                if cache_config.shared_kuzu_lock:
+                    self.redis_lock.acquire_lock()
+                    lock_acquired = True
                 if not self.connection:
-                    logger.debug("Reconnecting to Kuzu database...")
+                    logger.info("Reconnecting to Kuzu database...")
                     self._initialize_connection()
 
                 result = self.connection.execute(query, params)
@@ -225,12 +251,47 @@ class KuzuAdapter(GraphDBInterface):
                             val = val.as_py()
                         processed_rows.append(val)
                     rows.append(tuple(processed_rows))
+
                 return rows
             except Exception as e:
                 logger.error(f"Query execution failed: {str(e)}")
                 raise
+            finally:
+                if cache_config.shared_kuzu_lock and lock_acquired:
+                    try:
+                        self.close()
+                    finally:
+                        self.redis_lock.release_lock()
 
-        return await loop.run_in_executor(self.executor, blocking_query)
+        if cache_config.shared_kuzu_lock:
+            async with self._connection_change_lock:
+                self.open_connections += 1
+                logger.info(f"Open connections after open: {self.open_connections}")
+                try:
+                    result = blocking_query()
+                finally:
+                    self.open_connections -= 1
+                    logger.info(f"Open connections after close: {self.open_connections}")
+                return result
+        else:
+            result = await loop.run_in_executor(self.executor, blocking_query)
+            return result
+
+    def close(self):
+        if self.connection:
+            del self.connection
+            self.connection = None
+        if self.db:
+            del self.db
+            self.db = None
+        self._is_closed = True
+        logger.info("Kuzu database closed successfully")
+
+    def reopen(self):
+        if self._is_closed:
+            self._is_closed = False
+            self._initialize_connection()
+            logger.info("Kuzu database re-opened successfully")
 
     @asynccontextmanager
     async def get_session(self):
@@ -1557,44 +1618,6 @@ class KuzuAdapter(GraphDBInterface):
             logger.error(f"Failed to delete graph data: {e}")
             raise
 
-    async def clear_database(self) -> None:
-        """
-        Clear all data from the database by deleting the database files and reinitializing.
-
-        This method removes all files associated with the database and reinitializes the Kuzu
-        database structure, ensuring a completely empty state. It handles exceptions that might
-        occur during file deletions or initializations carefully.
-        """
-        try:
-            if self.connection:
-                self.connection = None
-            if self.db:
-                self.db.close()
-                self.db = None
-
-            db_dir = os.path.dirname(self.db_path)
-            db_name = os.path.basename(self.db_path)
-            file_storage = get_file_storage(db_dir)
-
-            if await file_storage.file_exists(db_name):
-                await file_storage.remove_all()
-                logger.info(f"Deleted Kuzu database files at {self.db_path}")
-
-            # Reinitialize the database
-            self._initialize_connection()
-            # Verify the database is empty
-            result = self.connection.execute("MATCH (n:Node) RETURN COUNT(n)")
-            count = result.get_next()[0] if result.has_next() else 0
-            if count > 0:
-                logger.warning(
-                    f"Database still contains {count} nodes after clearing, forcing deletion"
-                )
-                self.connection.execute("MATCH (n:Node) DETACH DELETE n")
-            logger.info("Database cleared successfully")
-        except Exception as e:
-            logger.error(f"Error during database clearing: {e}")
-            raise
-
     async def get_document_subgraph(self, data_id: str):
         """
         Get all nodes that should be deleted when removing a document.

cognee/infrastructure/databases/graph/neo4j_driver/adapter.py

@@ -87,6 +87,15 @@ class Neo4jAdapter(GraphDBInterface):
         async with self.driver.session(database=self.graph_database_name) as session:
             yield session
 
+    async def is_empty(self) -> bool:
+        query = """
+        RETURN EXISTS {
+        MATCH (n)
+        } AS node_exists;
+        """
+        query_result = await self.query(query)
+        return not query_result[0]["node_exists"]
+
     @deadlock_retry()
     async def query(
         self,

cognee/infrastructure/loaders/LoaderEngine.py

@@ -105,7 +105,6 @@ class LoaderEngine:
     async def load_file(
         self,
         file_path: str,
-        file_stream: Optional[Any],
         preferred_loaders: Optional[List[str]] = None,
         **kwargs,
     ):

cognee/modules/pipelines/operations/pipeline.py

@@ -35,6 +35,7 @@ async def run_pipeline(
     vector_db_config: dict = None,
     graph_db_config: dict = None,
     incremental_loading: bool = False,
+    data_per_batch: int = 20,
 ):
     validate_pipeline_tasks(tasks)
     await setup_and_check_environment(vector_db_config, graph_db_config)
@@ -50,6 +51,7 @@ async def run_pipeline(
             pipeline_name=pipeline_name,
             context={"dataset": dataset},
             incremental_loading=incremental_loading,
+            data_per_batch=data_per_batch,
         ):
             yield run_info
 
@@ -62,6 +64,7 @@ async def run_pipeline_per_dataset(
     pipeline_name: str = "custom_pipeline",
     context: dict = None,
     incremental_loading=False,
+    data_per_batch: int = 20,
 ):
     # Will only be used if ENABLE_BACKEND_ACCESS_CONTROL is set to True
     await set_database_global_context_variables(dataset.id, dataset.owner_id)
@@ -77,7 +80,7 @@ async def run_pipeline_per_dataset(
         return
 
     pipeline_run = run_tasks(
-        tasks, dataset.id, data, user, pipeline_name, context, incremental_loading
+        tasks, dataset.id, data, user, pipeline_name, context, incremental_loading, data_per_batch
     )
 
     async for pipeline_run_info in pipeline_run:

cognee/modules/pipelines/operations/run_tasks.py

@@ -24,7 +24,6 @@ from cognee.modules.pipelines.operations import (
     log_pipeline_run_complete,
     log_pipeline_run_error,
 )
-from .run_tasks_with_telemetry import run_tasks_with_telemetry
 from .run_tasks_data_item import run_tasks_data_item
 from ..tasks.task import Task
 
@@ -60,6 +59,7 @@ async def run_tasks(
     pipeline_name: str = "unknown_pipeline",
     context: dict = None,
     incremental_loading: bool = False,
+    data_per_batch: int = 20,
 ):
     if not user:
         user = await get_default_user()
@@ -89,7 +89,11 @@ async def run_tasks(
         if incremental_loading:
             data = await resolve_data_directories(data)
 
-        # Create async tasks per data item that will run the pipeline for the data item
+        # Create and gather batches of async tasks of data items that will run the pipeline for the data item
+        results = []
+        for start in range(0, len(data), data_per_batch):
+            data_batch = data[start : start + data_per_batch]
+
             data_item_tasks = [
                 asyncio.create_task(
                     run_tasks_data_item(
@@ -104,9 +108,10 @@ async def run_tasks(
                         incremental_loading,
                     )
                 )
-            for data_item in data
+                for data_item in data_batch
             ]
-        results = await asyncio.gather(*data_item_tasks)
+
+            results.extend(await asyncio.gather(*data_item_tasks))
 
         # Remove skipped data items from results
         results = [result for result in results if result]

cognee/modules/pipelines/operations/run_tasks_data_item.py

@@ -115,9 +115,8 @@ async def run_tasks_data_item_incremental(
             data_point = (
                 await session.execute(select(Data).filter(Data.id == data_id))
             ).scalar_one_or_none()
-            data_point.pipeline_status[pipeline_name] = {
-                str(dataset.id): DataItemStatus.DATA_ITEM_PROCESSING_COMPLETED
-            }
+            status_for_pipeline = data_point.pipeline_status.setdefault(pipeline_name, {})
+            status_for_pipeline[str(dataset.id)] = DataItemStatus.DATA_ITEM_PROCESSING_COMPLETED
             await session.merge(data_point)
             await session.commit()
 

cognee/tests/subprocesses/reader.py

@@ -0,0 +1,25 @@
+import asyncio
+import time
+from cognee.infrastructure.databases.graph.kuzu.adapter import KuzuAdapter
+
+# This will create the test.db if it doesn't exist
+
+
+async def main():
+    adapter = KuzuAdapter("test.db")
+    result = await adapter.query("MATCH (n:Node) RETURN COUNT(n)")
+    print(f"Reader: Found {result[0][0]} nodes")
+    result = await adapter.query("MATCH (n:Node) RETURN COUNT(n)")
+    print(f"Reader: Found {result[0][0]} nodes")
+    result = await adapter.query("MATCH (n:Node) RETURN COUNT(n)")
+    print(f"Reader: Found {result[0][0]} nodes")
+    result = await adapter.query("MATCH (n:Node) RETURN COUNT(n)")
+    print(f"Reader: Found {result[0][0]} nodes")
+    result = await adapter.query("MATCH (n:Node) RETURN COUNT(n)")
+    print(f"Reader: Found {result} nodes")
+    result = await adapter.query("MATCH (n:Node) RETURN COUNT(n)")
+    print(f"Reader: Found {result[0][0]} nodes")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

cognee/tests/subprocesses/simple_cognify_1.py

@@ -0,0 +1,31 @@
+import asyncio
+import cognee
+from cognee.shared.logging_utils import setup_logging, INFO
+from cognee.api.v1.search import SearchType
+
+
+async def main():
+    await cognee.cognify(datasets=["first_cognify_dataset"])
+
+    query_text = (
+        "Tell me what is in the context. Additionally write out 'FIRST_COGNIFY' before your answer"
+    )
+    search_results = await cognee.search(
+        query_type=SearchType.GRAPH_COMPLETION,
+        query_text=query_text,
+        datasets=["first_cognify_dataset"],
+    )
+
+    print("Search results:")
+    for result_text in search_results:
+        print(result_text)
+
+
+if __name__ == "__main__":
+    logger = setup_logging(log_level=INFO)
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+    try:
+        loop.run_until_complete(main())
+    finally:
+        loop.run_until_complete(loop.shutdown_asyncgens())

cognee/tests/subprocesses/simple_cognify_2.py

@@ -0,0 +1,31 @@
+import asyncio
+import cognee
+from cognee.shared.logging_utils import setup_logging, INFO
+from cognee.api.v1.search import SearchType
+
+
+async def main():
+    await cognee.cognify(datasets=["second_cognify_dataset"])
+
+    query_text = (
+        "Tell me what is in the context. Additionally write out 'SECOND_COGNIFY' before your answer"
+    )
+    search_results = await cognee.search(
+        query_type=SearchType.GRAPH_COMPLETION,
+        query_text=query_text,
+        datasets=["second_cognify_dataset"],
+    )
+
+    print("Search results:")
+    for result_text in search_results:
+        print(result_text)
+
+
+if __name__ == "__main__":
+    logger = setup_logging(log_level=INFO)
+    loop = asyncio.new_event_loop()
+    asyncio.set_event_loop(loop)
+    try:
+        loop.run_until_complete(main())
+    finally:
+        loop.run_until_complete(loop.shutdown_asyncgens())

cognee/tests/subprocesses/writer.py

@@ -0,0 +1,32 @@
+import asyncio
+import time
+import uuid
+from cognee.modules.data.processing.document_types import PdfDocument
+from cognee.infrastructure.databases.graph.kuzu.adapter import KuzuAdapter
+
+
+def create_node(name):
+    document = PdfDocument(
+        id=uuid.uuid4(),
+        name=name,
+        raw_data_location=name,
+        external_metadata="test_external_metadata",
+        mime_type="test_mime",
+    )
+    return document
+
+
+async def main():
+    adapter = KuzuAdapter("test.db")
+    nodes = [create_node(f"Node{i}") for i in range(5)]
+
+    print("Writer: Starting...")
+    await adapter.add_nodes(nodes)
+
+    print("writer finished...")
+
+    time.sleep(10)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

cognee/tests/test_concurrent_subprocess_access.py

@@ -0,0 +1,84 @@
+import os
+import asyncio
+import cognee
+import pathlib
+import subprocess
+
+from cognee.shared.logging_utils import get_logger
+
+logger = get_logger()
+
+"""
+Test: Redis-based Kùzu Locking Across Subprocesses
+
+This test ensures the Redis shared lock correctly serializes access to the Kùzu
+database when multiple subprocesses (writer/reader and cognify tasks) run in parallel.
+If this test fails, it indicates the locking mechanism is not properly handling
+concurrent subprocess access.
+"""
+
+
+async def concurrent_subprocess_access():
+    data_directory_path = str(
+        pathlib.Path(
+            os.path.join(pathlib.Path(__file__).parent, ".data_storage/concurrent_tasks")
+        ).resolve()
+    )
+    cognee_directory_path = str(
+        pathlib.Path(
+            os.path.join(pathlib.Path(__file__).parent, ".cognee_system/concurrent_tasks")
+        ).resolve()
+    )
+
+    subprocess_directory_path = str(
+        pathlib.Path(os.path.join(pathlib.Path(__file__).parent, "subprocesses/")).resolve()
+    )
+
+    writer_path = subprocess_directory_path + "/writer.py"
+    reader_path = subprocess_directory_path + "/reader.py"
+
+    cognee.config.data_root_directory(data_directory_path)
+    cognee.config.system_root_directory(cognee_directory_path)
+
+    await cognee.prune.prune_data()
+    await cognee.prune.prune_system(metadata=True)
+
+    writer_process = subprocess.Popen([os.sys.executable, str(writer_path)])
+
+    reader_process = subprocess.Popen([os.sys.executable, str(reader_path)])
+
+    # Wait for both processes to complete
+    writer_process.wait()
+    reader_process.wait()
+
+    logger.info("Basic write read subprocess example finished")
+
+    await cognee.prune.prune_data()
+    await cognee.prune.prune_system(metadata=True)
+
+    text = """
+            This is the text of the first cognify subprocess
+            """
+    await cognee.add(text, dataset_name="first_cognify_dataset")
+
+    text = """
+            This is the text of the second cognify subprocess
+            """
+    await cognee.add(text, dataset_name="second_cognify_dataset")
+
+    first_cognify_path = subprocess_directory_path + "/simple_cognify_1.py"
+    second_cognify_path = subprocess_directory_path + "/simple_cognify_2.py"
+
+    first_cognify_process = subprocess.Popen([os.sys.executable, str(first_cognify_path)])
+
+    second_cognify_process = subprocess.Popen([os.sys.executable, str(second_cognify_path)])
+
+    # Wait for both processes to complete
+    first_cognify_process.wait()
+    second_cognify_process.wait()
+
+    logger.info("Database concurrent subprocess example finished")
+
+
+if __name__ == "__main__":
+    asyncio.run(concurrent_subprocess_access())

cognee/tests/test_kuzu.py

@@ -47,10 +47,26 @@ async def main():
             pathlib.Path(__file__).parent, "test_data/Quantum_computers.txt"
         )
 
+        from cognee.infrastructure.databases.graph import get_graph_engine
+
+        graph_engine = await get_graph_engine()
+
+        is_empty = await graph_engine.is_empty()
+
+        assert is_empty, "Kuzu graph database is not empty"
+
         await cognee.add([explanation_file_path_quantum], dataset_name)
 
+        is_empty = await graph_engine.is_empty()
+
+        assert is_empty, "Kuzu graph database should be empty before cognify"
+
         await cognee.cognify([dataset_name])
 
+        is_empty = await graph_engine.is_empty()
+
+        assert not is_empty, "Kuzu graph database should not be empty"
+
         from cognee.infrastructure.databases.vector import get_vector_engine
 
         vector_engine = get_vector_engine()
@@ -114,11 +130,10 @@ async def main():
         assert not os.path.isdir(data_root_directory), "Local data files are not deleted"
 
         await cognee.prune.prune_system(metadata=True)
-        from cognee.infrastructure.databases.graph import get_graph_engine
 
-        graph_engine = await get_graph_engine()
-        nodes, edges = await graph_engine.get_graph_data()
-        assert len(nodes) == 0 and len(edges) == 0, "Kuzu graph database is not empty"
+        is_empty = await graph_engine.is_empty()
+
+        assert is_empty, "Kuzu graph database is not empty"
 
     finally:
         # Ensure cleanup even if tests fail

cognee/tests/test_memgraph.py

@@ -1,105 +0,0 @@
-import os
-
-import pathlib
-import cognee
-from cognee.infrastructure.files.storage import get_storage_config
-from cognee.modules.search.operations import get_history
-from cognee.modules.users.methods import get_default_user
-from cognee.shared.logging_utils import get_logger
-from cognee.modules.search.types import SearchType
-
-
-logger = get_logger()
-
-
-async def main():
-    cognee.config.set_graph_database_provider("memgraph")
-    data_directory_path = str(
-        pathlib.Path(
-            os.path.join(pathlib.Path(__file__).parent, ".data_storage/test_memgraph")
-        ).resolve()
-    )
-    cognee.config.data_root_directory(data_directory_path)
-    cognee_directory_path = str(
-        pathlib.Path(
-            os.path.join(pathlib.Path(__file__).parent, ".cognee_system/test_memgraph")
-        ).resolve()
-    )
-    cognee.config.system_root_directory(cognee_directory_path)
-
-    await cognee.prune.prune_data()
-    await cognee.prune.prune_system(metadata=True)
-
-    dataset_name = "cs_explanations"
-
-    explanation_file_path_nlp = os.path.join(
-        pathlib.Path(__file__).parent, "test_data/Natural_language_processing.txt"
-    )
-    await cognee.add([explanation_file_path_nlp], dataset_name)
-
-    explanation_file_path_quantum = os.path.join(
-        pathlib.Path(__file__).parent, "test_data/Quantum_computers.txt"
-    )
-
-    await cognee.add([explanation_file_path_quantum], dataset_name)
-
-    await cognee.cognify([dataset_name])
-
-    from cognee.infrastructure.databases.vector import get_vector_engine
-
-    vector_engine = get_vector_engine()
-    random_node = (await vector_engine.search("Entity_name", "Quantum computer"))[0]
-    random_node_name = random_node.payload["text"]
-
-    search_results = await cognee.search(
-        query_type=SearchType.GRAPH_COMPLETION, query_text=random_node_name
-    )
-    assert len(search_results) != 0, "The search results list is empty."
-    print("\n\nExtracted sentences are:\n")
-    for result in search_results:
-        print(f"{result}\n")
-
-    search_results = await cognee.search(query_type=SearchType.CHUNKS, query_text=random_node_name)
-    assert len(search_results) != 0, "The search results list is empty."
-    print("\n\nExtracted chunks are:\n")
-    for result in search_results:
-        print(f"{result}\n")
-
-    search_results = await cognee.search(
-        query_type=SearchType.SUMMARIES, query_text=random_node_name
-    )
-    assert len(search_results) != 0, "Query related summaries don't exist."
-    print("\nExtracted results are:\n")
-    for result in search_results:
-        print(f"{result}\n")
-
-    search_results = await cognee.search(
-        query_type=SearchType.NATURAL_LANGUAGE,
-        query_text=f"Find nodes connected to node with name {random_node_name}",
-    )
-    assert len(search_results) != 0, "Query related natural language don't exist."
-    print("\nExtracted results are:\n")
-    for result in search_results:
-        print(f"{result}\n")
-
-    user = await get_default_user()
-    history = await get_history(user.id)
-
-    assert len(history) == 8, "Search history is not correct."
-
-    await cognee.prune.prune_data()
-    data_root_directory = get_storage_config()["data_root_directory"]
-    assert not os.path.isdir(data_root_directory), "Local data files are not deleted"
-
-    await cognee.prune.prune_system(metadata=True)
-    from cognee.infrastructure.databases.graph import get_graph_engine
-
-    graph_engine = await get_graph_engine()
-    nodes, edges = await graph_engine.get_graph_data()
-    assert len(nodes) == 0 and len(edges) == 0, "Memgraph graph database is not empty"
-
-
-if __name__ == "__main__":
-    import asyncio
-
-    asyncio.run(main())

cognee/tests/test_neo4j.py

@@ -35,6 +35,14 @@ async def main():
     explanation_file_path_nlp = os.path.join(
         pathlib.Path(__file__).parent, "test_data/Natural_language_processing.txt"
     )
+    from cognee.infrastructure.databases.graph import get_graph_engine
+
+    graph_engine = await get_graph_engine()
+
+    is_empty = await graph_engine.is_empty()
+
+    assert is_empty, "Graph has to be empty"
+
     await cognee.add([explanation_file_path_nlp], dataset_name)
 
     explanation_file_path_quantum = os.path.join(
@@ -42,9 +50,16 @@ async def main():
     )
 
     await cognee.add([explanation_file_path_quantum], dataset_name)
+    is_empty = await graph_engine.is_empty()
+
+    assert is_empty, "Graph has to be empty before cognify"
 
     await cognee.cognify([dataset_name])
 
+    is_empty = await graph_engine.is_empty()
+
+    assert not is_empty, "Graph shouldn't be empty"
+
     from cognee.infrastructure.databases.vector import get_vector_engine
 
     vector_engine = get_vector_engine()
@@ -117,11 +132,8 @@ async def main():
     assert not os.path.isdir(data_root_directory), "Local data files are not deleted"
 
     await cognee.prune.prune_system(metadata=True)
-    from cognee.infrastructure.databases.graph import get_graph_engine
-
-    graph_engine = await get_graph_engine()
-    nodes, edges = await graph_engine.get_graph_data()
-    assert len(nodes) == 0 and len(edges) == 0, "Neo4j graph database is not empty"
+    is_empty = await graph_engine.is_empty()
+    assert is_empty, "Neo4j graph database is not empty"
 
 
 if __name__ == "__main__":

cognee/tests/unit/api/test_search.py

@@ -0,0 +1,21 @@
+import pytest
+import cognee
+
+
+@pytest.mark.asyncio
+async def test_empty_search_raises_SearchOnEmptyGraphError_on_empty_graph():
+    await cognee.prune.prune_data()
+    await cognee.prune.prune_system(metadata=True)
+    await cognee.add("Sample input")
+    result = await cognee.search("Sample query")
+    assert result == []
+
+
+@pytest.mark.asyncio
+async def test_empty_search_doesnt_raise_SearchOnEmptyGraphError():
+    await cognee.prune.prune_data()
+    await cognee.prune.prune_system(metadata=True)
+    await cognee.add("Sample input")
+    await cognee.cognify()
+    result = await cognee.search("Sample query")
+    assert result != []

cognee/tests/unit/infrastructure/databases/cache/test_cache_config.py

@@ -0,0 +1,87 @@
+"""Tests for cache configuration."""
+
+import pytest
+from cognee.infrastructure.databases.cache.config import CacheConfig, get_cache_config
+
+
+def test_cache_config_defaults():
+    """Test that CacheConfig has the correct default values."""
+    config = CacheConfig()
+
+    assert config.caching is False
+    assert config.shared_kuzu_lock is False
+    assert config.cache_host == "localhost"
+    assert config.cache_port == 6379
+    assert config.agentic_lock_expire == 240
+    assert config.agentic_lock_timeout == 300
+
+
+def test_cache_config_custom_values():
+    """Test that CacheConfig accepts custom values."""
+    config = CacheConfig(
+        caching=True,
+        shared_kuzu_lock=True,
+        cache_host="redis.example.com",
+        cache_port=6380,
+        agentic_lock_expire=120,
+        agentic_lock_timeout=180,
+    )
+
+    assert config.caching is True
+    assert config.shared_kuzu_lock is True
+    assert config.cache_host == "redis.example.com"
+    assert config.cache_port == 6380
+    assert config.agentic_lock_expire == 120
+    assert config.agentic_lock_timeout == 180
+
+
+def test_cache_config_to_dict():
+    """Test the to_dict method returns all configuration values."""
+    config = CacheConfig(
+        caching=True,
+        shared_kuzu_lock=True,
+        cache_host="test-host",
+        cache_port=7000,
+        agentic_lock_expire=100,
+        agentic_lock_timeout=200,
+    )
+
+    config_dict = config.to_dict()
+
+    assert config_dict == {
+        "caching": True,
+        "shared_kuzu_lock": True,
+        "cache_host": "test-host",
+        "cache_port": 7000,
+        "agentic_lock_expire": 100,
+        "agentic_lock_timeout": 200,
+    }
+
+
+def test_get_cache_config_singleton():
+    """Test that get_cache_config returns the same instance."""
+    config1 = get_cache_config()
+    config2 = get_cache_config()
+
+    assert config1 is config2
+
+
+def test_cache_config_extra_fields_allowed():
+    """Test that CacheConfig allows extra fields due to extra='allow'."""
+    config = CacheConfig(extra_field="extra_value", another_field=123)
+
+    assert hasattr(config, "extra_field")
+    assert config.extra_field == "extra_value"
+    assert hasattr(config, "another_field")
+    assert config.another_field == 123
+
+
+def test_cache_config_boolean_type_validation():
+    """Test that boolean fields accept various truthy/falsy values."""
+    config1 = CacheConfig(caching="true", shared_kuzu_lock="yes")
+    assert config1.caching is True
+    assert config1.shared_kuzu_lock is True
+
+    config2 = CacheConfig(caching="false", shared_kuzu_lock="no")
+    assert config2.caching is False
+    assert config2.shared_kuzu_lock is False

docker-compose.yml

@@ -129,6 +129,30 @@ services:
     networks:
       - cognee-network
 
+  redis:
+    image: redis:7-alpine
+    container_name: redis
+    profiles:
+      - redis
+    ports:
+      - "6379:6379"
+    networks:
+      - cognee-network
+    volumes:
+      - redis_data:/data
+    command: [ "redis-server", "--appendonly", "yes" ]
+
+
+  redisinsight:
+    image: redislabs/redisinsight:latest
+    container_name: redisinsight
+    restart: always
+    ports:
+      - "5540:5540"
+    networks:
+      - cognee-network
+
+
 networks:
   cognee-network:
     name: cognee-network
@@ -136,3 +160,4 @@ networks:
 volumes:
   chromadb_data:
   postgres_data:
+  redis_data:

notebooks/neptune-analytics-example.ipynb

@@ -83,16 +83,16 @@
    ]
   },
   {
-   "metadata": {},
    "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "import os\n",
     "import pathlib\n",
     "from cognee import config, add, cognify, search, SearchType, prune, visualize_graph\n",
     "from dotenv import load_dotenv"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -106,7 +106,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# load environment variables from file .env\n",
     "load_dotenv()\n",
@@ -145,9 +147,7 @@
     "        \"vector_db_url\": f\"neptune-graph://{graph_identifier}\",  # Neptune Analytics endpoint with the format neptune-graph://<GRAPH_ID>\n",
     "    }\n",
     ")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -159,19 +159,19 @@
    ]
   },
   {
-   "metadata": {},
    "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "# Prune data and system metadata before running, only if we want \"fresh\" state.\n",
     "await prune.prune_data()\n",
     "await prune.prune_system(metadata=True)"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
-   "metadata": {},
    "cell_type": "markdown",
+   "metadata": {},
    "source": [
     "## Setup data and cognify\n",
     "\n",
@@ -180,7 +180,9 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Add sample text to the dataset\n",
     "sample_text_1 = \"\"\"Neptune Analytics is a memory-optimized graph database engine for analytics. With Neptune\n",
@@ -205,9 +207,7 @@
     "\n",
     "# Cognify the text data.\n",
     "await cognify([dataset_name])"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -215,14 +215,16 @@
    "source": [
     "## Graph Memory visualization\n",
     "\n",
-    "Initialize Memgraph as a Graph Memory store and save to .artefacts/graph_visualization.html\n",
+    "Initialize Neptune as a Graph Memory store and save to .artefacts/graph_visualization.html\n",
     "\n",
     "![visualization](./neptune_analytics_demo.png)"
    ]
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Get a graphistry url (Register for a free account at https://www.graphistry.com)\n",
     "# url = await render_graph()\n",
@@ -235,9 +237,7 @@
     "    ).resolve()\n",
     ")\n",
     "await visualize_graph(graph_file_path)"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
    "cell_type": "markdown",
@@ -250,19 +250,19 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Completion query that uses graph data to form context.\n",
     "graph_completion = await search(query_text=\"What is Neptune Analytics?\", query_type=SearchType.GRAPH_COMPLETION)\n",
     "print(\"\\nGraph completion result is:\")\n",
     "print(graph_completion)"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
-   "metadata": {},
    "cell_type": "markdown",
+   "metadata": {},
    "source": [
     "## SEARCH: RAG Completion\n",
     "\n",
@@ -271,19 +271,19 @@
   },
   {
    "cell_type": "code",
+   "execution_count": null,
    "metadata": {},
+   "outputs": [],
    "source": [
     "# Completion query that uses document chunks to form context.\n",
     "rag_completion = await search(query_text=\"What is Neptune Analytics?\", query_type=SearchType.RAG_COMPLETION)\n",
     "print(\"\\nRAG Completion result is:\")\n",
     "print(rag_completion)"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
-   "metadata": {},
    "cell_type": "markdown",
+   "metadata": {},
    "source": [
     "## SEARCH: Graph Insights\n",
     "\n",
@@ -291,8 +291,10 @@
    ]
   },
   {
-   "metadata": {},
    "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "# Search graph insights\n",
     "insights_results = await search(query_text=\"Neptune Analytics\", query_type=SearchType.GRAPH_COMPLETION)\n",
@@ -302,13 +304,11 @@
     "    tgt_node = result[2].get(\"name\", result[2][\"type\"])\n",
     "    relationship = result[1].get(\"relationship_name\", \"__relationship__\")\n",
     "    print(f\"- {src_node} -[{relationship}]-> {tgt_node}\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
-   "metadata": {},
    "cell_type": "markdown",
+   "metadata": {},
    "source": [
     "## SEARCH: Entity Summaries\n",
     "\n",
@@ -316,8 +316,10 @@
    ]
   },
   {
-   "metadata": {},
    "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "# Query all summaries related to query.\n",
     "summaries = await search(query_text=\"Neptune Analytics\", query_type=SearchType.SUMMARIES)\n",
@@ -326,13 +328,11 @@
     "    type = summary[\"type\"]\n",
     "    text = summary[\"text\"]\n",
     "    print(f\"- {type}: {text}\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   },
   {
-   "metadata": {},
    "cell_type": "markdown",
+   "metadata": {},
    "source": [
     "## SEARCH: Chunks\n",
     "\n",
@@ -340,8 +340,10 @@
    ]
   },
   {
-   "metadata": {},
    "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
    "source": [
     "chunks = await search(query_text=\"Neptune Analytics\", query_type=SearchType.CHUNKS)\n",
     "print(\"\\nChunk results are:\")\n",
@@ -349,9 +351,7 @@
     "    type = chunk[\"type\"]\n",
     "    text = chunk[\"text\"]\n",
     "    print(f\"- {type}: {text}\")"
-   ],
-   "outputs": [],
-   "execution_count": null
+   ]
   }
  ],
  "metadata": {

pyproject.toml

@@ -7,7 +7,7 @@ authors = [
     { name = "Vasilije Markovic" },
     { name = "Boris Arzentar" },
 ]
-requires-python = ">=3.10,<=3.13"
+requires-python = ">=3.10,<3.14"
 readme = "README.md"
 license = "Apache-2.0"
 classifiers = [
@@ -67,10 +67,10 @@ distributed = [
 ]
 
 scraping = [
-    "tavily-python>=0.7.0",
+    "tavily-python>=0.7.12",
     "beautifulsoup4>=4.13.1",
     "playwright>=1.9.0",
-    "lxml>=4.9.3,<5.0.0",
+    "lxml>=4.9.3",
     "protego>=0.1",
     "APScheduler>=3.10.0,<=3.11.0"
 ]
@@ -104,7 +104,7 @@ chromadb = [
     "chromadb>=0.6,<0.7",
     "pypika==0.48.9",
 ]
-docs = ["unstructured[csv, doc, docx, epub, md, odt, org, ppt, pptx, rst, rtf, tsv, xlsx, pdf]>=0.18.1,<19"]
+docs = ["lxml<6.0.0", "unstructured[csv, doc, docx, epub, md, odt, org, ppt, pptx, rst, rtf, tsv, xlsx, pdf]>=0.18.1,<19"]
 codegraph = [
     "fastembed<=0.6.0 ; python_version < '3.13'",
     "transformers>=4.46.3,<5",
@@ -143,6 +143,7 @@ dev = [
     "mkdocstrings[python]>=0.26.2,<0.27",
 ]
 debug = ["debugpy>=1.8.9,<2.0.0"]
+redis = ["redis>=5.0.3,<6.0.0"]
 
 monitoring = ["sentry-sdk[fastapi]>=2.9.0,<3", "langfuse>=2.32.0,<3"]
 

uv.lock

@@ -1084,6 +1084,9 @@ postgres-binary = [
 posthog = [
     { name = "posthog" },
 ]
+redis = [
+    { name = "redis" },
+]
 scraping = [
     { name = "apscheduler" },
     { name = "beautifulsoup4" },
@@ -1174,6 +1177,7 @@ requires-dist = [
     { name = "python-magic-bin", marker = "sys_platform == 'win32'", specifier = "<0.5" },
     { name = "python-multipart", specifier = ">=0.0.20,<1.0.0" },
     { name = "rdflib", specifier = ">=7.1.4,<7.2.0" },
+    { name = "redis", marker = "extra == 'redis'", specifier = ">=5.0.3,<6.0.0" },
     { name = "ruff", marker = "extra == 'dev'", specifier = ">=0.9.2,<=0.13.1" },
     { name = "s3fs", extras = ["boto3"], marker = "extra == 'aws'", specifier = "==2025.3.2" },
     { name = "scikit-learn", marker = "extra == 'evals'", specifier = ">=1.6.1,<2" },
@@ -1195,7 +1199,7 @@ requires-dist = [
     { name = "uvicorn", specifier = ">=0.34.0,<1.0.0" },
     { name = "websockets", specifier = ">=15.0.1,<16.0.0" },
 ]
-provides-extras = ["api", "distributed", "scraping", "neo4j", "neptune", "postgres", "postgres-binary", "notebook", "langchain", "llama-index", "huggingface", "ollama", "mistral", "anthropic", "deepeval", "posthog", "groq", "chromadb", "docs", "codegraph", "evals", "graphiti", "aws", "dlt", "baml", "dev", "debug", "monitoring", "docling"]
+provides-extras = ["api", "distributed", "scraping", "neo4j", "neptune", "postgres", "postgres-binary", "notebook", "langchain", "llama-index", "huggingface", "ollama", "mistral", "anthropic", "deepeval", "posthog", "groq", "chromadb", "docs", "codegraph", "evals", "graphiti", "aws", "dlt", "baml", "dev", "debug", "redis", "monitoring", "docling"]
 
 [[package]]
 name = "colorama"
@@ -7379,6 +7383,19 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/f4/31/e9b6f04288dcd3fa60cb3179260d6dad81b92aef3063d679ac7d80a827ea/rdflib-7.1.4-py3-none-any.whl", hash = "sha256:72f4adb1990fa5241abd22ddaf36d7cafa5d91d9ff2ba13f3086d339b213d997", size = 565051, upload-time = "2025-03-29T02:22:44.987Z" },
 ]
 
+[[package]]
+name = "redis"
+version = "5.3.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "async-timeout", marker = "python_full_version < '3.11.3'" },
+    { name = "pyjwt" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/6a/cf/128b1b6d7086200c9f387bd4be9b2572a30b90745ef078bd8b235042dc9f/redis-5.3.1.tar.gz", hash = "sha256:ca49577a531ea64039b5a36db3d6cd1a0c7a60c34124d46924a45b956e8cf14c", size = 4626200, upload-time = "2025-07-25T08:06:27.778Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/7f/26/5c5fa0e83c3621db835cfc1f1d789b37e7fa99ed54423b5f519beb931aa7/redis-5.3.1-py3-none-any.whl", hash = "sha256:dc1909bd24669cc31b5f67a039700b16ec30571096c5f1f0d9d2324bff31af97", size = 272833, upload-time = "2025-07-25T08:06:26.317Z" },
+]
+
 [[package]]
 name = "referencing"
 version = "0.37.0"

working_dir_error_replication/run_subprocess_test.py

@@ -0,0 +1,31 @@
+"""
+Run writer and reader in separate subprocesses to test Kuzu locks.
+"""
+
+import subprocess
+import time
+import os
+
+
+def main():
+    print("=== Kuzu Subprocess Lock Test ===")
+    print("Starting writer and reader in separate subprocesses...")
+    print("Writer will hold the database lock, reader should block or fail\n")
+
+    start_time = time.time()
+
+    # Start writer subprocess
+    writer_process = subprocess.Popen([os.sys.executable, "writer.py"])
+
+    reader_process = subprocess.Popen([os.sys.executable, "reader.py"])
+
+    # Wait for both processes to complete
+    writer_process.wait()
+    reader_process.wait()
+
+    total_time = time.time() - start_time
+    print(f"\nTotal execution time: {total_time:.2f}s")
+
+
+if __name__ == "__main__":
+    main()