Update app.py

app.py

@@ -0,0 +1,55 @@
+import sys
+import os
+import re
+import shutil
+import time
+import fitz 
+import streamlit as st
+import nltk
+import tempfile
+import subprocess
+
+# Pin NLTK to version 3.9.1
+REQUIRED_NLTK_VERSION = "3.9.1"
+subprocess.run([sys.executable, "-m", "pip", "install", f"nltk=={REQUIRED_NLTK_VERSION}"])
+
+# Set up temporary directory for NLTK resources
+nltk_data_path = os.path.join(tempfile.gettempdir(), "nltk_data")
+os.makedirs(nltk_data_path, exist_ok=True)
+nltk.data.path.append(nltk_data_path)
+
+# Download 'punkt_tab' for compatibility
+try:
+    print("Ensuring NLTK 'punkt_tab' resource is downloaded...")
+    nltk.download("punkt_tab", download_dir=nltk_data_path)
+except Exception as e:
+    print(f"Error downloading NLTK 'punkt_tab': {e}")
+    raise e
+
+sys.path.append(os.path.abspath("."))
+from langchain.chains import ConversationalRetrievalChain
+from langchain.memory import ConversationBufferMemory
+from langchain.llms import OpenAI
+from langchain.document_loaders import UnstructuredPDFLoader
+from langchain.vectorstores import Chroma
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.text_splitter import NLTKTextSplitter
+from patent_downloader import PatentDownloader
+from langchain.document_loaders import PyMuPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+PERSISTED_DIRECTORY = tempfile.mkdtemp()
+
+# Fetch API key securely from the environment
+OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
+if not OPENAI_API_KEY:
+    st.error("Critical Error: OpenAI API key not found in the environment variables. Please configure it.")
+    st.stop()
+
+def check_poppler_installed():
+    if not shutil.which("pdfinfo"):
+        raise EnvironmentError(
+            "Poppler is not installed or not in PATH. Install 'poppler-utils' for PDF processing."
+        )
+
+check_poppler_installed()