Update app.py

app.py

@@ -0,0 +1,104 @@
+import os
+import requests
+from groq import Groq
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from PyPDF2 import PdfReader
+import streamlit as st
+from tempfile import NamedTemporaryFile
+
+# Initialize Groq client
+client = Groq(api_key=os.environ['GROQ_API_KEY'])
+
+# Function to extract text from a PDF
+def extract_text_from_pdf(pdf_file_path):
+    pdf_reader = PdfReader(pdf_file_path)
+    text = ""
+    for page in pdf_reader.pages:
+        page_text = page.extract_text()
+        if page_text:
+            text += page_text
+    return text
+
+# Function to split text into chunks
+def chunk_text(text, chunk_size=500, chunk_overlap=50):
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=chunk_size, chunk_overlap=chunk_overlap
+    )
+    return text_splitter.split_text(text)
+
+# Function to create embeddings and store them in FAISS
+def create_embeddings_and_store(chunks, vector_db=None):
+    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+    if vector_db is None:
+        vector_db = FAISS.from_texts(chunks, embedding=embeddings)
+    else:
+        vector_db.add_texts(chunks)
+    return vector_db
+
+# Function to query the vector database and interact with Groq
+def query_vector_db(query, vector_db):
+    docs = vector_db.similarity_search(query, k=3)
+    context = "\n".join([doc.page_content for doc in docs])
+    chat_completion = client.chat.completions.create(
+        messages=[
+            {"role": "system", "content": f"Use the following context:\n{context}"},
+            {"role": "user", "content": query},
+        ],
+        model="llama3-8b-8192",
+    )
+    return chat_completion.choices[0].message.content
+
+# Function to convert Google Drive view link to downloadable link
+def get_direct_download_link(view_url):
+    if "drive.google.com/file/d/" in view_url:
+        file_id = view_url.split("/file/d/")[1].split("/")[0]
+        return f"https://drive.google.com/uc?export=download&id={file_id}"
+    return None
+
+# Function to download and save a PDF from a URL
+def download_pdf_from_url(url):
+    direct_url = get_direct_download_link(url)
+    if not direct_url:
+        return None
+    response = requests.get(direct_url)
+    if response.status_code == 200:
+        temp_file = NamedTemporaryFile(delete=False, suffix=".pdf")
+        temp_file.write(response.content)
+        temp_file.close()
+        return temp_file.name
+    else:
+        return None
+
+# Streamlit app
+st.title("RAG-Based QA on Google Drive PDFs")
+
+# Only fetch from provided links
+doc_links = [
+    "https://drive.google.com/file/d/1YWX-RYxgtcKO1QETnz1N3rboZUhRZwcH/view?usp=sharing",
+    "https://drive.google.com/file/d/1JPf0XvDhn8QoDOlZDrxCOpu4WzKFESNz/view?usp=sharing",
+]
+
+vector_db = None
+
+# Process Google Drive documents
+for idx, link in enumerate(doc_links):
+    st.write(f"📄 Fetching and processing PDF from Link {idx + 1}...")
+    pdf_path = download_pdf_from_url(link)
+    if pdf_path:
+        text = extract_text_from_pdf(pdf_path)
+        chunks = chunk_text(text)
+        vector_db = create_embeddings_and_store(chunks, vector_db=vector_db)
+        st.success(f"✅ Processed document {idx + 1}")
+    else:
+        st.error(f"❌ Failed to download or process PDF from Link {idx + 1}")
+
+# User query input
+user_query = st.text_input("🔍 Enter your query:")
+if user_query and vector_db:
+    response = query_vector_db(user_query, vector_db)
+    st.subheader("💬 Response from LLM:")
+    st.write(response)
+elif user_query:
+    st.warning("⚠️ No documents processed to query.")