!conda install -n pa \
    pytorch \
    torchvision \
    torchaudio \
    cpuonly \
    -c pytorch \
    -c conda-forge \
    --yes
%pip install -U ipywidgets
%pip install -U requests
%pip install -U llama-index
%pip install -U llama-index-embeddings-huggingface
%pip install -U llama-index-llms-groq
%pip install -U groq
%pip install -U gradio

import os
import platform
import subprocess
import requests


def install_tesseract():
    """
    Installs Tesseract OCR based on the operating system.
    """
    os_name = platform.system()
    if os_name == "Linux":
        print("Detected Linux. Installing Tesseract using apt-get...")
        subprocess.run(["sudo", "apt-get", "update"], check=True)
        subprocess.run(["sudo", "apt-get", "install", "-y", "tesseract-ocr"], check=True)
    elif os_name == "Darwin":
        print("Detected macOS. Installing Tesseract using Homebrew...")
        subprocess.run(["brew", "install", "tesseract"], check=True)
    elif os_name == "Windows":
        tesseract_installer_url = "https://github.com/UB-Mannheim/tesseract/releases/download/v5.4.0.20240606/tesseract-ocr-w64-setup-5.4.0.20240606.exe"
        installer_path = "tesseract-ocr-w64-setup-5.4.0.20240606.exe"
        response = requests.get(tesseract_installer_url)
        with open(installer_path, "wb") as file:
            file.write(response.content)
        tesseract_path = r"C:\Program Files\Tesseract-OCR"
        os.environ["PATH"] += os.pathsep + tesseract_path
        try:
            result = subprocess.run(["tesseract", "--version"], check=True, capture_output=True, text=True)
            print(result.stdout)
        except subprocess.CalledProcessError as e:
            print(f"Error running Tesseract: {e}")
    else:
        print(f"Unsupported OS: {os_name}")


install_tesseract()

import webbrowser

url = "https://www.ilovepdf.com/ocr-pdf"
webbrowser.open_new(url)

import os
from llama_index.core import (
    Settings,
    VectorStoreIndex,
    SimpleDirectoryReader,
    StorageContext,
    load_index_from_storage
)
from llama_index.embeddings.huggingface import HuggingFaceEmbedding
from llama_index.core.node_parser import SentenceSplitter
from llama_index.llms.groq import Groq
import gradio as gr