import streamlit as st
import pickle
import pandas as pd
import requests
from bs4 import BeautifulSoup
import re

# === Load model & vectorizer ===
@st.cache_resource
def load_model(model_choice):
    with open(f'model/{model_choice}_model.pkl', 'rb') as f:
        model = pickle.load(f)
    with open('model/tfidf_vectorizer.pkl', 'rb') as f:
        vectorizer = pickle.load(f)
    return model, vectorizer

label_names = ['susu', 'kacang', 'telur', 'makanan_laut', 'gandum']

# === Text Cleaning ===
def clean_text(text):
    text = re.sub(r'[^\w\s]', ' ', text.lower())
    text = re.sub(r'\d+', '', text)
    return text.strip()

# === Scraping Cookpad ===
def scrape_ingredients(url):
    try:
        headers = {'User-Agent': 'Mozilla/5.0'}
        r = requests.get(url, headers=headers)
        soup = BeautifulSoup(r.content, 'html.parser')
        ingredients_div = soup.find('div', id='ingredients')
        if ingredients_div:
            return ingredients_div.get_text(separator=' ')
    except:
        pass
    return None

# === UI ===
st.title("🍲 Deteksi Alergen dari Resep Cookpad")

# Hapus SVM dari pilihan
model_choice = st.selectbox("🔍 Pilih model:", options=["KNN", "RF"])
model, vectorizer = load_model(model_choice)

input_mode = st.radio("Pilih mode input:", ["Teks Manual", "Link Cookpad.com"])

if input_mode == "Teks Manual":
    user_input = st.text_area("📝 Masukkan teks bahan makanan:")
else:
    url_input = st.text_input("🔗 Masukkan URL resep dari cookpad.com:")
    if url_input:
        scraped = scrape_ingredients(url_input)
        if scraped:
            user_input = scraped
            st.success("✅ Bahan berhasil diambil dari URL!")
            st.text_area("📋 Bahan yang diambil:", value=user_input, height=200)
        else:
            user_input = ""
            st.error("Gagal mengambil data dari URL. Pastikan URL valid dan dari cookpad.com.")

threshold = st.slider("🎚 Threshold prediksi (default 0.5):", 0.0, 1.0, 0.5)

if st.button("🚀 Prediksi"):
    if user_input.strip():
        cleaned_text = clean_text(user_input)
        user_vector = vectorizer.transform([cleaned_text])

        if hasattr(model, "predict_proba"):
            user_proba = model.predict_proba(user_vector)
            user_proba = [p[0][1] for p in user_proba]  # probability of class 1
        else:
            user_proba = model.predict(user_vector)[0]
            user_proba = [float(val) for val in user_proba]

        st.subheader(f"📊 Hasil Prediksi Alergen ({model_choice}):")
        for label, proba in zip(label_names, user_proba):
            status = "✅ Ada" if proba >= threshold else "❌ Tidak Ada"
            st.write(f"- **{label}**: {status} ({proba:.2f})")
    else:
        st.warning("❗ Masukkan teks terlebih dahulu atau URL valid.")