Update app.py

app.py

@@ -1,77 +1,57 @@
-# app.py
 import streamlit as st
-import requests
-from bs4 import BeautifulSoup
-from openai import OpenAI  # New-style OpenAI client for Groq
-
-# Set your Groq API key and base URL
-client = OpenAI(
-    api_key="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",  # Replace with your actual Groq API key
-    base_url="https://api.groq.com/openai/v1"
-)
-
-def extract_text_from_url(url):
-    try:
-        response = requests.get(url)
-        soup = BeautifulSoup(response.text, 'html.parser')
-        paragraphs = soup.find_all('p')
-        text = ' '.join([p.get_text() for p in paragraphs])
-        return text
-    except Exception as e:
-        return f"Error fetching {url}: {e}"
-
-def generate_blog(content, keywords):
-    prompt = f"""
-You are a professional SEO blog writer. 
-Based on the following combined content, generate a completely new, attractive, and SEO-optimized blog.
-Please naturally incorporate the following keywords: {', '.join(keywords)}.
-
-Content:
-{content}
-
-Write the new blog post:
-"""
-    response = client.chat.completions.create(
-        model="llama3-70b-8192",  # LLaMA 3 via Groq
-        messages=[
-            {"role": "system", "content": "You are an expert SEO content writer."},
-            {"role": "user", "content": prompt}
-        ],
+from langchain_community.document_loaders import PyPDFLoader
+from langchain_community.vectorstores import FAISS
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.chains import RetrievalQA
+from langchain_community.llms import Groq
+import os
+
+# Set up page
+st.set_page_config(page_title="Chat with PDF")
+
+# Sidebar: enter Groq API key
+st.sidebar.title("Configuration")
+groq_api_key = st.sidebar.text_input("Enter your Groq API Key", type="password")
+
+# Main heading
+st.title("📄 PDF Chatbot using LangChain + FAISS + Groq")
+st.markdown("This app answers your questions based on a predefined PDF.")
+
+# Path to your preloaded PDF
+PDF_PATH = "src/acca.pdf"
+
+# Process PDF and create retriever (cached)
+@st.cache_resource
+def load_retriever(pdf_path):
+    loader = PyPDFLoader(pdf_path)
+    documents = loader.load_and_split()
+    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+    db = FAISS.from_documents(documents, embeddings)
+    return db.as_retriever()
+
+# Validate API key
+if not groq_api_key:
+    st.warning("Please enter your Groq API key.")
+else:
+    retriever = load_retriever(PDF_PATH)
+    
+    # Initialize Groq LLM
+    llm = Groq(
+        model="llama3-8b-8192",
+        api_key=groq_api_key,
         temperature=0.7,
-        max_tokens=1500
     )
-    return response.choices[0].message.content
-
-def main():
-    st.title("📝 Pro SEO Blog Writer")
-
-    st.subheader("Enter three blog URLs:")
-    url1 = st.text_input("Blog URL 1")
-    url2 = st.text_input("Blog URL 2")
-    url3 = st.text_input("Blog URL 3")
 
-    st.subheader("Enter Target Keywords (comma separated):")
-    keywords_input = st.text_input("Example: AI, machine learning, future technology")
-
-    if st.button("Generate New Blog"):
-        if url1 and url2 and url3 and keywords_input:
-            with st.spinner("Extracting content and generating blog..."):
-                content1 = extract_text_from_url(url1)
-                content2 = extract_text_from_url(url2)
-                content3 = extract_text_from_url(url3)
-
-                combined_content = content1 + "\n\n" + content2 + "\n\n" + content3
-                keywords = [kw.strip() for kw in keywords_input.split(",")]
-
-                new_blog = generate_blog(combined_content, keywords)
-
-            st.success("✅ Blog generated successfully!")
-            st.subheader("Generated Blog:")
-            st.write(new_blog)
-
-            st.download_button("Download Blog as TXT", data=new_blog, file_name="seo_blog.txt")
-        else:
-            st.warning("Please fill in all fields.")
+    # QA chain
+    qa_chain = RetrievalQA.from_chain_type(
+        llm=llm,
+        chain_type="stuff",
+        retriever=retriever
+    )
 
-if __name__ == "__main__":
-    main()
+    # Input field for questions
+    user_input = st.text_input("Ask a question about the PDF:")
+    if user_input:
+        with st.spinner("Generating response..."):
+            answer = qa_chain.run(user_input)
+        st.success(answer)