Update src/streamlit_app.py

src/streamlit_app.py

@@ -1,25 +1,36 @@
 import streamlit as st
+import zipfile
+import os
 from langchain_community.vectorstores import Chroma
 from langchain_community.embeddings import HuggingFaceEmbeddings
 
 st.set_page_config(page_title="RAG Search", page_icon="🔍")
 
-DB_PATH = "path_to_your_existing_db"
-query = st.text_input("Enter your query:", "what is verilog")
+# --- 1️⃣ Unzip chroma_db.zip if not already extracted ---
+ZIP_PATH = os.path.join(os.path.dirname(__file__), "..", "chroma_db.zip")
+DB_PATH = os.path.join(os.path.dirname(__file__), "..", "chroma_db")
 
-# FIX 🔧: Force model to run on CPU
+if not os.path.exists(DB_PATH):
+    st.info("📦 Extracting Chroma DB for first-time setup...")
+    with zipfile.ZipFile(ZIP_PATH, "r") as zip_ref:
+        zip_ref.extractall(DB_PATH)
+    st.success("✅ Database extracted successfully!")
+
+# --- 2️⃣ Initialize embedding model ---
 embeddings = HuggingFaceEmbeddings(
     model_name="mixedbread-ai/mxbai-embed-large-v1",
-    model_kwargs={"device": "cpu"}
+    model_kwargs={"device": "cpu"}  # Force CPU for Hugging Face Spaces
 )
 
-vectordb = Chroma(persist_directory=DB_PATH)
+# --- 3️⃣ Load Chroma database ---
+vectordb = Chroma(persist_directory=DB_PATH, embedding_function=embeddings)
+
+# --- 4️⃣ Query input & results ---
+query = st.text_input("Enter your query:", "What is SystemVerilog interface?")
 
 if st.button("Search"):
     st.write("🔎 Searching your local vector database...")
-    query_vector = embeddings.embed_query(query)
-    results = vectordb.similarity_search_by_vector(query_vector, k=3)
-
+    results = vectordb.similarity_search(query, k=3)
     if results:
         for i, doc in enumerate(results):
             st.subheader(f"Result {i+1}")
@@ -27,5 +38,4 @@ if st.button("Search"):
             st.caption(doc.metadata)
             st.markdown("---")
     else:
-        st.warning("No matching results found.")
-
+        st.warning("⚠️ No results found.")