app.py

import PyPDF2
import gradio as gr
from smolagents import CodeAgent, HfApiModel, tool
from tools.final_answer import FinalAnswerTool

# Define the PDF text extraction tool
@tool
def extract_text_from_pdf(pdf_path: str) -> str:
    """Extracts text from a given PDF file.
    Args:
        pdf_path: The path to the PDF file.
    """
    try:
        with open(pdf_path, "rb") as file:
            reader = PyPDF2.PdfReader(file)
            text = "\n".join([page.extract_text() for page in reader.pages if page.extract_text()])
        return text if text else "No text found in the PDF."
    except Exception as e:
        return f"Error extracting text from PDF: {str(e)}"

# Define the PDF summarization tool
@tool
def summarize_text(text: str) -> str:
    """Summarizes the extracted text using the AI model.
    Args:
        text: The extracted text from the PDF.
    """
    max_input_tokens = 15000  # ✅ Adjust input limit to stay within model constraints
    
    # Truncate text if it's too long
    truncated_text = text[:max_input_tokens]

    prompt = f"Summarize the following document:\n\n{truncated_text}"
    
    # ✅ Correct input format for HfApiModel
    response = model.__call__([{"role": "user", "content": prompt}])

    return response

# Initialize AI Model
model = HfApiModel(
    max_tokens=1024,  # Reduced token limit for summaries
    temperature=0.5,
    model_id='Qwen/Qwen2.5-Coder-32B-Instruct',
    custom_role_conversions=None,
)

# Initialize Final Answer Tool
final_answer = FinalAnswerTool()

# Create AI Agent with PDF extraction and summarization tools
agent = CodeAgent(
    model=model,
    tools=[final_answer, extract_text_from_pdf, summarize_text],  # Added summarization tool
    max_steps=6,
    verbosity_level=1,
    grammar=None,
    planning_interval=None,
    name="PDF Summary Agent",
    description="An AI agent that extracts and summarizes text from PDFs.",
    prompt_templates=None
)

# Define Gradio function to handle PDF upload and summarization
def extract_and_summarize_pdf(file_path):
    extracted_text = extract_text_from_pdf(file_path)
    if "Error" in extracted_text:
        return extracted_text, "No summary available."
    
    summary = summarize_text(extracted_text)
    return extracted_text, summary

# Gradio UI Setup
with gr.Blocks() as ui:
    gr.Markdown("# 📄 PDF Summarizer Agent")
    gr.Markdown("### Upload a PDF file, and the agent will extract and summarize its content.")

    with gr.Row():
        pdf_input = gr.File(type="filepath", label="Upload PDF")  # ✅ Fixed type
        output_text = gr.Textbox(label="Extracted Text", interactive=False)
        summary_output = gr.Textbox(label="Summary", interactive=False)

    extract_button = gr.Button("Extract & Summarize")
    extract_button.click(extract_and_summarize_pdf, inputs=pdf_input, outputs=[output_text, summary_output])

# Launch Gradio UI
ui.launch()