app.py

import os
import re
import cv2
import time
import numpy as np
import pandas as pd
import requests
import streamlit as st
from PIL import Image
from paddleocr import PaddleOCR

# ==============================
# KONFIGURASI APLIKASI
# ==============================
st.set_page_config(
    page_title="Nutri-Grade Label Detection",
    page_icon="🥗",
    layout="wide",
    initial_sidebar_state="collapsed"
)

# API OpenRouter
OPENROUTER_API_KEY = "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"
OPENROUTER_BASE_URL = "https://openrouter.ai/api/v1"


# ==============================
# FUNGSI UTAMA
# ==============================

@st.cache_resource
def initialize_ocr():
    """Inisialisasi model PaddleOCR (CPU)."""
    try:
        return PaddleOCR(lang='en', use_angle_cls=True)
    except Exception as e:
        st.error(f"Gagal inisialisasi OCR: {e}")
        return None


def parse_numeric_value(text: str) -> float:
    """Mengubah string menjadi float, hanya menyisakan angka/desimal."""
    cleaned = re.sub(r"[^\d\.\-]", "", str(text))
    if cleaned in ['', '.', '-']:
        return 0.0
    try:
        return float(cleaned)
    except ValueError:
        return 0.0


def get_nutrition_advice(serving_size, sugar_norm, fat_norm, sugar_grade, fat_grade, final_grade):
    """Memanggil API OpenRouter untuk menghasilkan saran nutrisi singkat."""
    prompt = f"""
    Anda adalah ahli gizi dari Indonesia yang ramah.
    - Takaran Saji: {serving_size} g/ml
    - Gula (per 100): {sugar_norm:.2f} g (Grade {sugar_grade.replace('Grade ', '')})
    - Lemak Jenuh (per 100): {fat_norm:.2f} g (Grade {fat_grade.replace('Grade ', '')})
    - Grade Akhir: {final_grade.replace('Grade ', '')}
    
    Berikan saran nutrisi singkat 50-80 kata, fokus pada dampak kesehatan dan tips praktis.
    """
    headers = {"Authorization": f"Bearer {OPENROUTER_API_KEY}", "Content-Type": "application/json"}
    payload = {
        "model": "mistralai/mistral-7b-instruct:free",
        "messages": [{"role": "user", "content": prompt}],
        "max_tokens": 250,
        "temperature": 0.7,
    }
    try:
        r = requests.post(f"{OPENROUTER_BASE_URL}/chat/completions", headers=headers, json=payload, timeout=30)
        r.raise_for_status()
        return r.json()["choices"][0]["message"]["content"].strip()
    except Exception as e:
        return f"Error: {e}"


def get_grade_from_value(value, thresholds):
    """Menentukan grade berdasarkan ambang batas."""
    if value <= thresholds["A"]:
        return "Grade A"
    if value <= thresholds["B"]:
        return "Grade B"
    if value <= thresholds["C"]:
        return "Grade C"
    return "Grade D"


def get_grade_color(grade):
    """Mengembalikan warna background & teks untuk tiap grade."""
    return {
        "Grade A": ("#2ecc71", "white"),
        "Grade B": ("#f1c40f", "black"),
        "Grade C": ("#e67e22", "white"),
        "Grade D": ("#e74c3c", "white")
    }.get(grade, ("#bdc3c7", "black"))


def reset_state():
    """Reset session_state agar analisis bisa diulang."""
    for key in ['ocr_done', 'data', 'calculated', 'calc']:
        st.session_state.pop(key, None)


# ==============================
# UI APLIKASI
# ==============================

# Inisialisasi OCR
ocr_engine = initialize_ocr()
if ocr_engine is None:
    st.error("Model OCR tidak tersedia.")
    st.stop()

st.title("🥗 Nutri-Grade Detection & Grade Calculator")
st.caption("Analisis gizi produk berdasarkan standar Nutri-Grade Singapura.")

with st.expander("📋 Petunjuk Penggunaan"):
    st.markdown("""
    1. Upload gambar (JPG/PNG).
    2. Klik **Analisis OCR**.
    3. Koreksi hasil jika perlu.
    4. Klik **Hitung Grade**.
    """)

# --- Step 1: Upload ---
st.header("1. Upload Gambar")
file = st.file_uploader("Pilih gambar tabel gizi", type=["jpg", "jpeg", "png"], on_change=reset_state)

if file:
    arr = np.frombuffer(file.read(), np.uint8)
    img = cv2.imdecode(arr, cv2.IMREAD_COLOR)
    st.image(cv2.cvtColor(img, cv2.COLOR_BGR2RGB), width=300)

    if st.button("Analisis OCR"):
        with st.spinner("Mendeteksi teks..."):
            res = ocr_engine.ocr(img)

        texts = [ln[1][0] for ln in (res[0] if res else [])]
        full_text = " ".join(texts).lower()

        patterns = {
            'serving': r"(takaran saj[i|a]|serving size)[^\d]*(\d+\.?\d*)",
            'sugar': r"(gula|sugar)[^\d]*(\d+\.?\d*)",
            'fat': r"(lemak jenuh|saturated fat)[^\d]*(\d+\.?\d*)"
        }

        data = {}
        for key, pattern in patterns.items():
            match = re.search(pattern, full_text)
            if match:
                data[key] = match.group(2)

        st.session_state.data = data
        st.session_state.ocr_done = True
        st.success("OCR selesai!")
        st.rerun()

# --- Step 2: Koreksi & Hitung ---
if st.session_state.get('ocr_done'):
    st.header("2. Koreksi & Hitung Grade")
    d = st.session_state.data

    with st.form("form2"):
        serving = st.text_input("Takaran Saji (g/ml)", value=d.get('serving', '100'))
        sugar = st.text_input("Gula (g)", value=d.get('sugar', '0'))
        fat = st.text_input("Lemak Jenuh (g)", value=d.get('fat', '0'))
        ok = st.form_submit_button("Hitung Grade")

    if ok:
        sv = parse_numeric_value(serving)
        sg = parse_numeric_value(sugar)
        fg = parse_numeric_value(fat)

        sugar_per100 = (sg / sv) * 100 if sv > 0 else 0
        fat_per100 = (fg / sv) * 100 if sv > 0 else 0

        st.session_state.calc = {'sv': sv, 'sp': sugar_per100, 'fp': fat_per100}
        st.session_state.calculated = True

# --- Step 3: Tampilkan Hasil ---
if st.session_state.get('calculated'):
    c = st.session_state.calc

    gs = get_grade_from_value(c['sp'], {"A": 1.0, "B": 5.0, "C": 10.0})
    gf = get_grade_from_value(c['fp'], {"A": 0.7, "B": 1.2, "C": 2.8})

    final_grade = max(gs, gf, key=lambda x: ['Grade A', 'Grade B', 'Grade C', 'Grade D'].index(x))

    st.header("3. Hasil Grading")
    cols = st.columns(3)

    def show(col, title, value, unit, grade):
        bg, text_color = get_grade_color(grade)
        col.markdown(
            f"<div style='background:{bg};padding:10px;border-radius:8px;text-align:center;color:{text_color};'>"
            f"<strong>{title}</strong><p>{value:.2f} {unit}</p><h4>{grade}</h4></div>",
            unsafe_allow_html=True
        )

    show(cols[0], "Gula", c['sp'], "g/100ml", gs)
    show(cols[1], "Lemak Jenuh", c['fp'], "g/100ml", gf)
    show(cols[2], "Grade Akhir", 0, "", final_grade)

    st.divider()
    st.header("4. Saran Nutrisi AI")

    with st.spinner("Meminta AI..."):
        advice = get_nutrition_advice(c['sv'], c['sp'], c['fp'], gs, gf, final_grade)

    st.info(advice)

# --- Footer ---
st.markdown("---")
st.markdown("<p style='text-align:center;'>Nutri-Grade App v2.1 &copy; 2024</p>", unsafe_allow_html=True)