dataframe.py

import requests  # For making HTTP requests
import pandas as pd
import streamlit as st
from io import BytesIO
from bs4 import BeautifulSoup  # Add this import for BeautifulSoup


# Assuming fetch_data() is already defined, call this function to fetch the data
@st.cache_data(ttl=3600)
def fetch_data():
    # URL of the website to scrape
    url = "https://www.ireland.ie/en/india/newdelhi/services/visas/processing-times-and-decisions/"
    headers = {
        "User-Agent": (
            "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 "
            "(KHTML, like Gecko) Chrome/114.0.0.0 Safari/537.36"
        )
    }

    # Fetch the webpage
    response = requests.get(url, headers=headers)
    if response.status_code != 200:
        st.error("Failed to fetch the webpage. Please try again later.")
        return None

    # Parse the HTML to find the .ods link
    soup = BeautifulSoup(response.content, "html.parser")
    file_url = None
    for link in soup.find_all("a"):
        if "Visa decisions made from 1 January 2025" in link.get_text():
            file_url = link.get("href")
            if not file_url.startswith("http"):
                file_url = requests.compat.urljoin(url, file_url)
            break

    if not file_url:
        st.error("Could not find the visa decisions file link on the website.")
        return None

    # Fetch the .ods file
    ods_response = requests.get(file_url, headers=headers)
    if ods_response.status_code != 200:
        st.error("Failed to download the visa decisions file.")
        return None

    # Process the .ods file
    ods_file = BytesIO(ods_response.content)
    df = pd.read_excel(ods_file, engine="odf")
    
    # Drop unnecessary columns
    df.dropna(how="all", inplace=True)  # Drop rows with all NaN values
    df.reset_index(drop=True, inplace=True)

    # Keep only the first two columns
    if len(df.columns) > 2:
        df = df.iloc[:, :2]  # Keep only the first two columns
    
    # Rename columns
    if len(df.columns) == 2:
        df.columns = ["Application Number", "Decision"]
    else:
        st.error("Insufficient data columns detected.")
        return None

    df["Application Number"] = df["Application Number"].astype(str)

    # Store the dataframe as a global variable for future use
    return df

# Now, define precomputed_df to be used in other scripts
precomputed_df = fetch_data()  # Precompute the dataframe