utils.py

from typing import Dict, Any
import streamlit as st
from streamlit.runtime.uploaded_file_manager import UploadedFile
from typing import Dict, Union
import os
import time
from config import *

def validate_file(uploaded_file: UploadedFile) -> Dict[str, Union[bool, str, float]]:
    if not uploaded_file:
        return {"valid": False, "error": "No file uploaded"}
    
    # Check file extension
    file_extension = os.path.splitext(uploaded_file.name)[1].lower()
    if file_extension not in SUPPORTED_FORMATS:
        return {
            "valid": False, 
            "error": f"Unsupported format. Please upload: {', '.join(SUPPORTED_FORMATS)}"
        }
    
    # Check file size
    file_size_mb = uploaded_file.size / (1024 * 1024)
    if file_size_mb > MAX_FILE_SIZE_MB:
        return {
            "valid": False,
            "error": f"File too large. Maximum size: {MAX_FILE_SIZE_MB}MB"
        }
    
    return {"valid": True, "size_mb": round(file_size_mb, 2)}

def display_text_stats(stats: Dict[str, int]):
    if not stats:
        return
    
    col1, col2, col3, col4 = st.columns(4)
    
    with col1:
        st.metric("Characters", f"{stats.get('characters', 0):,}")
    
    with col2:
        st.metric("Words", f"{stats.get('words', 0):,}")
    
    with col3:
        st.metric("Sentences", f"{stats.get('sentences', 0):,}")
    
    with col4:
        st.metric("Est. Pages", stats.get('pages_estimated', 0))

def display_summary_results(result: Dict[str, Any]):
    if not result.get('success', False):
        st.error(f"{result.get('error', 'Unknown error')}")
        return
    
    # Display metrics
    col1, col2, col3, col4 = st.columns(4)
    
    with col1:
        st.metric("Original Words", f"{result['original_length']:,}")
    
    with col2:
        st.metric("Summary Words", f"{result['summary_length']:,}")
    
    with col3:
        st.metric("Compression Ratio", f"{result['compression_ratio']}%")
    
    with col4:
        st.metric("Chunks Processed", result['chunks_processed'])
    
    # Display performance metrics if available
    if 'total_time' in result:
        st.info(f"**Processing completed in {result['total_time']:.2f} seconds**")
    
    # Display summary
    st.subheader("Summary")
    st.write(result['summary'])
    
    # Download option
    st.download_button(
        label="📥 Download Summary",
        data=result['summary'],
        file_name="pdf_summary.txt",
        mime="text/plain"
    )

def create_summary_report(original_text: str, result: Dict[str, Any], 
                         filename: str) -> str:
    if not result.get('success'):
        return ""
    
    report = f"""
# PDF Summary Report

## Document Information
- **Original File:** {filename}
- **Processing Date:** {time.strftime('%Y-%m-%d %H:%M:%S')}
- **Original Text Length:** {result['original_length']} words
- **Summary Length:** {result['summary_length']} words
- **Compression Ratio:** {result['compression_ratio']}%
- **Chunks Processed:** {result['chunks_processed']}

## Summary
{result['summary']}

--- 
*Generated by PDF Summarizer using distilbart-cnn-12-6*
"""
    
    return report

def display_model_info():
    with st.expander("About the AI Model"):
        st.markdown("""
        **Model:** `sshleifer/distilbart-cnn-12-6`
        
        - **Type:** Distilled BART model for summarization
        - **Optimized for:** News articles and general text summarization
        - **Max Input Length:** ~1024 tokens per chunk
        - **Advantages:** Faster inference, good quality summaries
        - **Processing:** Text is automatically chunked for longer documents
        """)

def setup_page_config():
    st.set_page_config(
        page_title=PAGE_TITLE,
        page_icon=PAGE_ICON,
        layout=LAYOUT,
        initial_sidebar_state="collapsed"
    )