feat: Enhance application signal handling and update dependencies

- Added signal handling for graceful shutdown on SIGINT (Ctrl+C) in `app.py`.
- Introduced `db_persist` configuration option in the `Configuration` class to manage database persistence.
- Updated `DatastoreManager` to utilize `pathvalidate` for sanitizing file paths and filenames.
- Included `pathvalidate` as a new dependency in `pyproject.toml` and updated `uv.lock` accordingly.

app.py

@@ -1,6 +1,7 @@
 from pstuts_rag.configuration import Configuration
 import asyncio
 from typing import cast
+import signal
 
 import chainlit as cl
 from dotenv import load_dotenv
@@ -299,5 +300,10 @@ async def main(input_message: cl.Message):
 
 
 if __name__ == "__main__":
-    # This will start the Chainlit app when run directly
+
+    def handle_sigint(signum, frame):
+        print("SIGINT received (Ctrl+C), exiting...")
+        sys.exit(0)
+
+    signal.signal(signal.SIGINT, handle_sigint)
     pass

pstuts_rag/pstuts_rag/configuration.py

@@ -70,6 +70,8 @@ class Configuration:
 
     search_permission: str = str(os.environ.get("EVA_SEARCH_PERMISSION", "no"))
 
+    db_persist: str | None = os.environ.get("EVA_DB_PERSIST", None)
+
     thread_id: str = ""
 
     @classmethod

pstuts_rag/pstuts_rag/datastore.py

@@ -1,4 +1,5 @@
 import asyncio
+import atexit
 import json
 import glob
 import aiofiles
@@ -24,6 +25,7 @@ from qdrant_client import QdrantClient
 from qdrant_client.http.models import Distance, VectorParams
 from qdrant_client.models import PointStruct
 from pstuts_rag.utils import get_embeddings_api, flatten, batch
+from pathvalidate import sanitize_filename, sanitize_filepath
 
 
 class DatastoreManager:
@@ -56,7 +58,7 @@ class DatastoreManager:
     def __init__(
         self,
         embeddings: Optional[Embeddings] = None,
-        qdrant_client: QdrantClient = QdrantClient(location=":memory:"),
+        qdrant_client: QdrantClient | None = None,
         name: str = str(object=uuid.uuid4()),
         config: Configuration = Configuration(),
     ) -> None:
@@ -76,7 +78,39 @@ class DatastoreManager:
             self.embeddings = embeddings
 
         self.name = name if name else config.eva_workflow_name
+
+        if qdrant_client is None:
+
+            try:
+                if (
+                    config.db_persist
+                    and isinstance(config.db_persist, str)
+                    and len(config.db_persist) > 0
+                ):
+                    qdrant_path = Path(
+                        sanitize_filepath(config.db_persist)
+                    ) / sanitize_filename(config.embedding_model)
+                    logging.info(
+                        "Persisting the datastore to: %s",
+                        str(qdrant_path),
+                    )
+
+                    qdrant_path.mkdir(parents=True, exist_ok=True)
+
+                    qdrant_client = QdrantClient(path=str(qdrant_path))
+            except (OSError, ValueError) as e:
+                logging.error(
+                    "Persistence aborted, exception occurred: %s: %s",
+                    type(e).__name__,
+                    str(e),
+                )
+            finally:
+                if qdrant_client is None:
+                    qdrant_client = QdrantClient(location=":memory:")
+
         self.qdrant_client = qdrant_client
+        atexit.register(qdrant_client.close)
+
         self.loading_complete = asyncio.Event()
         self._completion_callbacks = []
 

pyproject.toml

@@ -51,6 +51,7 @@ dependencies = [
     "langgraph-cli[inmem]>=0.1.55",
     "langchain-tavily>=0.2.0",
     "beautifulsoup4>=4.13.4",
+    "pathvalidate>=3.2.3",
 ]
 authors = [{ name = "Marko Budisic", email = "mbudisic@gmail.com" }]
 license = "MIT"

uv.lock

@@ -3509,6 +3509,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/cc/20/ff623b09d963f88bfde16306a54e12ee5ea43e9b597108672ff3a408aad6/pathspec-0.12.1-py3-none-any.whl", hash = "sha256:a0d503e138a4c123b27490a4f7beda6a01c6f288df0e4a8b79c7eb0dc7b4cc08", size = 31191 },
 ]
 
+[[package]]
+name = "pathvalidate"
+version = "3.2.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/92/87/c7a2f51cc62df0495acb0ed2533a7c74cc895e569a1b020ee5f6e9fa4e21/pathvalidate-3.2.3.tar.gz", hash = "sha256:59b5b9278e30382d6d213497623043ebe63f10e29055be4419a9c04c721739cb", size = 61717 }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/50/14/c5a0e1a947909810fc4c043b84cac472b70e438148d34f5393be1bac663f/pathvalidate-3.2.3-py3-none-any.whl", hash = "sha256:5eaf0562e345d4b6d0c0239d0f690c3bd84d2a9a3c4c73b99ea667401b27bee1", size = 24130 },
+]
+
 [[package]]
 name = "pexpect"
 version = "4.9.0"
@@ -3766,6 +3775,7 @@ dependencies = [
     { name = "nest-asyncio" },
     { name = "numpy" },
     { name = "pandas" },
+    { name = "pathvalidate" },
     { name = "pyarrow" },
     { name = "python-dotenv" },
     { name = "qdrant-client" },
@@ -3838,6 +3848,7 @@ requires-dist = [
     { name = "nest-asyncio", specifier = ">=1.5.6" },
     { name = "numpy", specifier = ">=2.2.2" },
     { name = "pandas", specifier = ">=2.0.0" },
+    { name = "pathvalidate", specifier = ">=3.2.3" },
     { name = "pyarrow", specifier = ">=19.0.0" },
     { name = "pylint-venv", marker = "extra == 'dev'", specifier = ">=3.0.4" },
     { name = "pytest", marker = "extra == 'dev'", specifier = ">=7.0.0" },