Solution compiled successfully. Ready to publish.: update src/utils.py

src/utils.py

@@ -1,19 +1,20 @@
-import os
 import pathlib
 from typing import List
 
 from langchain_community.document_loaders import TextLoader
 from langchain_text_splitters import RecursiveCharacterTextSplitter
-from langchain_chroma import Chroma
+from langchain_qdrant import QdrantVectorStore
 from langchain_ollama import OllamaEmbeddings
 
-CHROMA_PATH = pathlib.Path("./chroma_faq")
-CHROMA_PATH.mkdir(parents=True, exist_ok=True)
+# Qdrant collection name
+COLLECTION_NAME = "faq"
+# Qdrant server URL (default local)
+QDRANT_URL = "http://localhost:6333"
 
 
-def load_faq_to_chroma(md_dir: str = "data") -> None:
-    """Load all .md files from md_dir into a Chroma vector store.
-    The store is persisted at CHROMA_PATH.
+def load_faq_to_qdrant(md_dir: str = "data") -> None:
+    """Load all .md files from md_dir into a Qdrant vector store.
+    The store is created/updated at COLLECTION_NAME.
     """
     loader = TextLoader
     all_docs = []
@@ -24,12 +25,22 @@ def load_faq_to_chroma(md_dir: str = "data") -> None:
     splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
     docs = splitter.split_documents(all_docs)
     embeddings = OllamaEmbeddings(model="nomic-embed-text")
-    Chroma.from_documents(docs, embeddings, persist_directory=str(CHROMA_PATH))
+    # Create or update Qdrant collection
+    QdrantVectorStore.from_documents(
+        docs,
+        embeddings,
+        collection_name=COLLECTION_NAME,
+        url=QDRANT_URL,
+    )
 
 
 def search_course_docs(query: str, k: int = 3) -> List[str]:
-    """Return top k document snippets from the persisted Chroma store."""
+    """Return top k document snippets from the persisted Qdrant store."""
     embeddings = OllamaEmbeddings(model="nomic-embed-text")
-    chroma = Chroma(persist_directory=str(CHROMA_PATH), embedding_function=embeddings)
-    results = chroma.similarity_search(query, k=k)
+    store = QdrantVectorStore(
+        collection_name=COLLECTION_NAME,
+        url=QDRANT_URL,
+        embedding_function=embeddings,
+    )
+    results = store.similarity_search(query, k=k)
     return [doc.page_content for doc in results]