rag-chat

.gitignore

@@ -3,3 +3,4 @@ venv
 __pycache__
 .vscode
 corpus
+

app.py

@@ -4,11 +4,6 @@ from langgraph.agents.rag_agent.graph import graph as rag_graph
 from fastapi import Request
 from fastapi.middleware.cors import CORSMiddleware
 
-from langchain_core.documents import Document
-
-from utils.create_vectordb import create_chroma_db_and_document,query_chroma_db
-
-
 
 
 
@@ -37,63 +32,12 @@ async def summarize(request: Request):
     notes = data.get("notes")
     return  graph.invoke({"urls": urls, "codes": codes, "notes": notes})
 
-
-@app.post("/save_summary")
-async def save_summary(request: Request):
-    data = await request.json()
-    summary = data.get("summary", "")
-    post_id = data.get("post_id", None)
-    title = data.get("title", "")
-    category = data.get("category", "") 
-    tags = data.get("tags", [])
-    references = data.get("references", [])
-    
-    page_content = f"""
-    Title: {title}
-    Category: {category}
-    Tags: {', '.join(tags)}
-    Summary: {summary}
-    """
-    
-    document = Document(
-        page_content=page_content,
-        id = str(post_id)
-        
-    )
-    
-    is_added = create_chroma_db_and_document(document)
-    
-    if not is_added:
-        return {"error": "Failed to save summary to the database." , "status": "error"}
-    
-    return {"message": "Summary saved successfully." , "status": "success"}
-
-@app.post("/summaries")
-async def get_summaries(request: Request):
-  
-    data = await request.json()
-    print(data)
-    query = data.get("query" , "")
-    
-    print(f"Query received: {query}")
-    results = query_chroma_db(query=query)
-    return results
-    
-    
-    
-    
 @app.post("/chat")
 async def chat(request: Request):
     data = await request.json()
-    
-    print(f"Chat request data: {data}")
-  
     user_input = data.get("message", "")
     chat_history = data.get("chat_history", [])
     
-    print(f"User input: {user_input}")
-    print(f"Chat history: {chat_history}")
-    
     # Invoke the RAG chatbot graph
     result = rag_graph.invoke({
         "user_input": user_input,

utils/create_vectordb.py

@@ -54,7 +54,7 @@ def split_documents(documents, chunk_size=1000, chunk_overlap=200):
     
     return splits
 
-def create_chroma_db_and_document(document, collection_name="corpus_collection", db_dir=DB_DIR):
+def create_chroma_db(documents, collection_name="corpus_collection", db_dir=DB_DIR):
     """Create a Chroma vector database from documents."""
     # Initialize the Gemini embedding function
     gemini_ef = embedding_functions.GoogleGenerativeAiEmbeddingFunction(
@@ -75,25 +75,17 @@ def create_chroma_db_and_document(document, collection_name="corpus_collection",
             embedding_function=gemini_ef
         )
         print(f"Created new collection: {collection_name}")
-        
-    
-    try:
     
+    # Add documents to collection
+    for i, doc in enumerate(documents):
         collection.add(
-            documents = [document.page_content],
-            ids = [document.id]
+            documents=[doc.page_content],
+            metadatas=[doc.metadata],
+            ids=[f"doc_{i}"]
         )
-        
-        print("Document added to collection successfully.")
-        return True
-    
-    except Exception as e:
-        print(f"Error adding document to collection: {e}")
-        
-        return False
-        
-        
- 
+    
+    print(f"Added {len(documents)} documents to collection {collection_name}")
+    return collection
 
 def query_chroma_db(query: str, collection_name="corpus_collection", n_results=5, db_dir=DB_DIR):
     """Query the Chroma vector database."""
@@ -144,10 +136,6 @@ def main():
         print(f"Source: {metadata.get('source', 'Unknown')}")
     
     print("\nVector database creation and testing complete!")
-    
-
-
-
 
 if __name__ == "__main__":
     main()