app.py

import streamlit as st
import pdfplumber
import requests
from io import BytesIO
import pandas as pd
import re

# Streamlit UI
st.title("📜 Find Information From Noticeboard")
st.write("Provide at least one user input (Name or Form Number).")

# Google Drive links mapped to meaningful names
google_drive_files = {
    "CCN Updated Attendance": "https://drive.google.com/file/d/17OqE5oSlXHlPlIzu7TLhnQXRWsE6TxQ1/view?usp=sharing",
      
}

# User inputs
name = st.text_input("Enter Name:", placeholder="Enter the full name to search for (e.g., John Doe).")
form_no = st.text_input("Enter Form Number:", placeholder="Enter the form number as it appears in the document.")

# Function to download PDFs from Google Drive links
def download_pdf_from_drive(link):
    try:
        file_id = link.split("/")[-2]  # Extract the file ID from the link
        download_url = f"https://drive.google.com/uc?id={file_id}&export=download"
        response = requests.get(download_url, stream=True)
        if response.status_code == 200:
            return BytesIO(response.content)
        else:
            st.warning(f"⚠ Failed to download file: {link} (Status code: {response.status_code})")
            return None
    except Exception as e:
        st.warning(f"⚠ Error: {e}")
        return None

# Function to extract column headers dynamically
def detect_column_headers(pdf_file):
    with pdfplumber.open(pdf_file) as pdf:
        for page in pdf.pages[:3]:  # Check first 3 pages for headers
            table = page.extract_table()
            if table and len(table) > 1 and table[0] is not None:
                return table[0]  # First row is header
            text = page.extract_text()
            if text:
                lines = text.split("\n")
                for line in lines:
                    columns = re.split(r'\s{2,}|\t', line.strip())
                    if len(columns) > 2:
                        return columns
    return None

# Function to extract relevant rows dynamically
def extract_relevant_rows(pdf_file, search_terms, file_name, detected_headers):
    relevant_rows = []
    with pdfplumber.open(pdf_file) as pdf:
        for page_num, page in enumerate(pdf.pages, start=1):
            table = page.extract_table()
            if table and len(table) > 1:
                for row in table[1:]:
                    if row and any(term.lower() in " ".join(map(str, row)).lower() for term in search_terms):
                        row_dict = {"File Name": file_name, "Page": page_num}
                        row_dict.update({detected_headers[i]: row[i] if i < len(row) else None for i in range(len(detected_headers))})
                        relevant_rows.append(row_dict)
            else:
                text = page.extract_text()
                if text:
                    lines = text.split("\n")
                    for line in lines:
                        for term in search_terms:
                            if term.lower() in line.lower():
                                columns = re.split(r'\s{2,}|\t', line.strip())
                                row_dict = {"File Name": file_name, "Page": page_num}
                                row_dict.update({detected_headers[i]: columns[i] if i < len(columns) else None for i in range(len(detected_headers))})
                                relevant_rows.append(row_dict)
    return relevant_rows

# Process the request when the button is clicked
if st.button("🔍 Find Information"):
    if not (name or form_no):
        st.error("❌ Please provide at least one input (Name or Form Number).")
    else:
        search_terms = [term for term in [name, form_no] if term]
        all_relevant_rows = []
        
        for file_name, link in google_drive_files.items():
            pdf_file = download_pdf_from_drive(link)
            if pdf_file:
                detected_headers = detect_column_headers(pdf_file)
                if not detected_headers:
                    st.warning(f"⚠ Unable to detect headers in {file_name}.")
                    continue
                
                relevant_rows = extract_relevant_rows(pdf_file, search_terms, file_name, detected_headers)
                if relevant_rows:
                    all_relevant_rows.extend(relevant_rows)
        
        if all_relevant_rows:
            st.success("✅ Relevant information found:")
            df = pd.DataFrame(all_relevant_rows).fillna("-")  # Fill NaN values for better readability
            st.dataframe(df)
        else:
            st.warning("⚠ No matching information found in the provided files.")

# Footer
st.markdown("---")
st.markdown(
    "<p style='text-align: center; font-size: 14px;'>Designed by: <b>Engr. Makhdoom Muhammad Naeem Hashmi</b></p>",
    unsafe_allow_html=True
)