init. project

rag-web-ui/backend/app/services/vector_store/__init__.py

@@ -0,0 +1,11 @@
+from .base import BaseVectorStore
+from .chroma import ChromaVectorStore
+from .qdrant import QdrantStore
+from .factory import VectorStoreFactory
+
+__all__ = [
+    'BaseVectorStore',
+    'ChromaVectorStore',
+    'QdrantStore',
+    'VectorStoreFactory'
+] 

rag-web-ui/backend/app/services/vector_store/base.py

@@ -0,0 +1,42 @@
+from abc import ABC, abstractmethod
+from typing import List, Optional, Dict, Any
+from langchain_core.documents import Document
+from langchain_core.embeddings import Embeddings
+
+class BaseVectorStore(ABC):
+    """Abstract base class for vector store implementations"""
+    
+    @abstractmethod
+    def __init__(self, collection_name: str, embedding_function: Embeddings, **kwargs):
+        """Initialize the vector store"""
+        pass
+    
+    @abstractmethod
+    def add_documents(self, documents: List[Document]) -> None:
+        """Add documents to the vector store"""
+        pass
+    
+    @abstractmethod
+    def delete(self, ids: List[str]) -> None:
+        """Delete documents from the vector store"""
+        pass
+    
+    @abstractmethod
+    def as_retriever(self, **kwargs: Any):
+        """Return a retriever interface for the vector store"""
+        pass
+    
+    @abstractmethod
+    def similarity_search(self, query: str, k: int = 4, **kwargs: Any) -> List[Document]:
+        """Search for similar documents"""
+        pass
+    
+    @abstractmethod
+    def similarity_search_with_score(self, query: str, k: int = 4, **kwargs: Any) -> List[Document]:
+        """Search for similar documents with score"""
+        pass
+
+    @abstractmethod
+    def delete_collection(self) -> None:
+        """Delete the entire collection"""
+        pass 

rag-web-ui/backend/app/services/vector_store/chroma.py

@@ -0,0 +1,47 @@
+from typing import List, Any
+from langchain_core.documents import Document
+from langchain_core.embeddings import Embeddings
+from langchain_chroma import Chroma
+import chromadb 
+from app.core.config import settings
+
+from .base import BaseVectorStore
+
+class ChromaVectorStore(BaseVectorStore):
+    """Chroma vector store implementation"""
+    
+    def __init__(self, collection_name: str, embedding_function: Embeddings, **kwargs):
+        """Initialize Chroma vector store"""
+        chroma_client = chromadb.HttpClient(
+            host=settings.CHROMA_DB_HOST,
+            port=settings.CHROMA_DB_PORT,
+        )
+        
+        self._store = Chroma(
+            client=chroma_client,
+            collection_name=collection_name,
+            embedding_function=embedding_function,
+        )
+    def add_documents(self, documents: List[Document]) -> None:
+        """Add documents to Chroma"""
+        self._store.add_documents(documents)
+    
+    def delete(self, ids: List[str]) -> None:
+        """Delete documents from Chroma"""
+        self._store.delete(ids)
+    
+    def as_retriever(self, **kwargs: Any):
+        """Return a retriever interface"""
+        return self._store.as_retriever(**kwargs)
+    
+    def similarity_search(self, query: str, k: int = 4, **kwargs: Any) -> List[Document]:
+        """Search for similar documents in Chroma"""
+        return self._store.similarity_search(query, k=k, **kwargs)
+    
+    def similarity_search_with_score(self, query: str, k: int = 4, **kwargs: Any) -> List[Document]:
+        """Search for similar documents in Chroma with score"""
+        return self._store.similarity_search_with_score(query, k=k, **kwargs)
+
+    def delete_collection(self) -> None:
+        """Delete the entire collection"""
+        self._store._client.delete_collection(self._store._collection.name) 

rag-web-ui/backend/app/services/vector_store/factory.py

@@ -0,0 +1,59 @@
+from typing import Dict, Type, Any
+from langchain_core.embeddings import Embeddings
+
+from .base import BaseVectorStore
+from .chroma import ChromaVectorStore
+from .qdrant import QdrantStore
+
+class VectorStoreFactory:
+    """Factory for creating vector store instances"""
+    
+    _stores: Dict[str, Type[BaseVectorStore]] = {
+        'chroma': ChromaVectorStore,
+        'qdrant': QdrantStore
+    }
+    
+    @classmethod
+    def create(
+        cls,
+        store_type: str,
+        collection_name: str,
+        embedding_function: Embeddings,
+        **kwargs: Any
+    ) -> BaseVectorStore:
+        """Create a vector store instance
+        
+        Args:
+            store_type: Type of vector store ('chroma', 'qdrant', etc.)
+            collection_name: Name of the collection
+            embedding_function: Embedding function to use
+            **kwargs: Additional arguments for specific vector store implementations
+            
+        Returns:
+            An instance of the requested vector store
+            
+        Raises:
+            ValueError: If store_type is not supported
+        """
+        store_class = cls._stores.get(store_type.lower())
+        if not store_class:
+            raise ValueError(
+                f"Unsupported vector store type: {store_type}. "
+                f"Supported types are: {', '.join(cls._stores.keys())}"
+            )
+        
+        return store_class(
+            collection_name=collection_name,
+            embedding_function=embedding_function,
+            **kwargs
+        )
+    
+    @classmethod
+    def register_store(cls, name: str, store_class: Type[BaseVectorStore]) -> None:
+        """Register a new vector store implementation
+        
+        Args:
+            name: Name of the vector store type
+            store_class: Vector store class implementation
+        """
+        cls._stores[name.lower()] = store_class 

rag-web-ui/backend/app/services/vector_store/qdrant.py

@@ -0,0 +1,43 @@
+from typing import List, Any
+from langchain_core.documents import Document
+from langchain_core.embeddings import Embeddings
+from langchain_community.vectorstores import Qdrant
+from app.core.config import settings
+
+from .base import BaseVectorStore
+
+class QdrantStore(BaseVectorStore):
+    """Qdrant vector store implementation"""
+    
+    def __init__(self, collection_name: str, embedding_function: Embeddings, **kwargs):
+        """Initialize Qdrant vector store"""
+        self._store = Qdrant(
+            collection_name=collection_name,
+            embeddings=embedding_function,
+            url=settings.QDRANT_URL,
+            prefer_grpc=settings.QDRANT_PREFER_GRPC
+        )
+    
+    def add_documents(self, documents: List[Document]) -> None:
+        """Add documents to Qdrant"""
+        self._store.add_documents(documents)
+    
+    def delete(self, ids: List[str]) -> None:
+        """Delete documents from Qdrant"""
+        self._store.delete(ids)
+    
+    def as_retriever(self, **kwargs: Any):
+        """Return a retriever interface"""
+        return self._store.as_retriever(**kwargs)
+    
+    def similarity_search(self, query: str, k: int = 4, **kwargs: Any) -> List[Document]:
+        """Search for similar documents in Qdrant"""
+        return self._store.similarity_search(query, k=k, **kwargs)
+    
+    def similarity_search_with_score(self, query: str, k: int = 4, **kwargs: Any) -> List[Document]:
+        """Search for similar documents in Qdrant with score"""
+        return self._store.similarity_search_with_score(query, k=k, **kwargs)
+
+    def delete_collection(self) -> None:
+        """Delete the entire collection"""
+        self._store._client.delete_collection(self._store._collection_name)