import sys import json import datetime import os # --- 1. GLOBAL LOGGING SETUP --- # We setup logging immediately to catch import errors or early failures LOG_FILE = os.path.join(os.path.dirname(__file__), "debug_python.log") def log_debug(msg): try: with open(LOG_FILE, "a") as f: f.write(f"[{datetime.datetime.now()}] {msg}\n") except: pass # Print log location for debugging visibility (Removed for Production Compliance) # print(f"DEBUG_LOG_PATH: {os.path.abspath(LOG_FILE)}") log_debug("--- Script Started ---") # --- 2. ROBUST IMPORTS --- try: import mysql.connector from mysql.connector import Error log_debug("MySQL Connector imported successfully.") except ImportError as e: log_debug(f"CRITICAL: Failed to import mysql.connector: {e}") print(json.dumps({"status": "error", "message": "Missing mysql-connector-python"})) sys.exit(1) # Import ML libraries (Optional - Fallback if missing) ML_AVAILABLE = False # [OPTIMIZATION] Disabled heavy libraries for speed (Reduces load time from 7s to 0.5s) # try: # import joblib # import pandas as pd # ML_AVAILABLE = True # log_debug("ML libraries (joblib, pandas) imported successfully.") # except ImportError as e: # log_debug(f"WARNING: ML libraries missing ({e}). Switching to Rule-Based Mode.") # --- 3. DATABASE CONNECTION --- def create_db_connection(): connection = None try: connection = mysql.connector.connect( host='localhost', user='root', password='', database='ayurpredict' ) if connection.is_connected(): log_debug("Database connected successfully.") except Error as e: log_debug(f"Database connection FAILED: {e}") return connection # --- 4. MAIN LOGIC --- def get_recommendations(user_id): log_debug(f"Processing for User ID: {user_id}") conn = create_db_connection() if not conn: print(json.dumps({"status": "error", "message": "Database connection failed"})) return try: cursor = conn.cursor(dictionary=True) # A. Fetch Scores query_scores = """ SELECT vata_score, pitta_score, kapha_score FROM dosha_scores WHERE user_id = %s ORDER BY checkin_date DESC LIMIT 1 """ cursor.execute(query_scores, (user_id,)) row = cursor.fetchone() if not row: log_debug("No scores found for user.") print(json.dumps({"status": "error", "message": "No check-in data found"})) return log_debug(f"Scores Found: Vata={row['vata_score']}, Pitta={row['pitta_score']}, Kapha={row['kapha_score']}") scores = { "Vata": row['vata_score'], "Pitta": row['pitta_score'], "Kapha": row['kapha_score'] } # B. Define Dominant Dosha (AI or Rule-Based) dominant_dosha = None if ML_AVAILABLE: model_path = os.path.join(os.path.dirname(__file__), 'dosha_model.pkl') if os.path.exists(model_path): try: model = joblib.load(model_path) input_df = pd.DataFrame([[scores['Vata'], scores['Pitta'], scores['Kapha']]], columns=['Vata', 'Pitta', 'Kapha']) prediction = model.predict(input_df)[0] log_debug(f"AI Prediction: {prediction}") if "-" in prediction: dominant_dosha = prediction.split("-")[0] else: dominant_dosha = prediction except Exception as ml_e: log_debug(f"AI Prediction Error: {ml_e}") else: log_debug("Model file (dosha_model.pkl) not found.") if not dominant_dosha: # Fallback sorted_scores = sorted(scores.items(), key=lambda item: item[1], reverse=True) dominant_dosha = sorted_scores[0][0] log_debug(f"Fallback Dominant: {dominant_dosha}") # Determine Lowest for support sorted_scores_all = sorted(scores.items(), key=lambda item: item[1], reverse=True) lowest_dosha = sorted_scores_all[2][0] # C. Recommendation Logic (Dynamic) targets = [] primary_dom = dominant_dosha dom_score = scores.get(primary_dom, 0) depleted_score = scores.get(lowest_dosha, 0) log_debug(f"Logic Check: DomScore={dom_score}, DepletedScore={depleted_score}") if dom_score > 60: log_debug("Condition: High Imbalance (>60)") targets.append({"dosha": primary_dom, "limit": 8}) if depleted_score <= 30: log_debug("Condition: Adding Depleted Support (<=30)") targets.append({"dosha": lowest_dosha, "limit": 4}) else: log_debug("Condition: Balanced (<=60)") targets.append({"dosha": primary_dom, "limit": 10}) # D. Fetch Data final_food = [] final_yoga = [] final_life = [] final_herbs = [] categories = ['food', 'yoga', 'lifestyle'] # base_image_url = "http://10.0.2.2/ayurpredict/images/herbs/" # Android Emulator base_image_url = "http://10.65.241.223/ayurpredict/images/herbs/" # User's Real IP for target in targets: d_name = target['dosha'] limit_val = target['limit'] # Tips for cat in categories: q_recs = "SELECT content FROM recommendations WHERE dosha = %s AND category = %s LIMIT %s" cursor.execute(q_recs, (d_name, cat, limit_val)) rows = cursor.fetchall() log_debug(f"Fetched {len(rows)} {cat} items for {d_name}") for r in rows: if cat == 'food': final_food.append(r['content']) elif cat == 'yoga': final_yoga.append(r['content']) elif cat == 'lifestyle': final_life.append(r['content']) # Herbs (Detailed) q_herbs = "SELECT name, description, image_filename, benefits, usage_dosage, usage_preparation, usage_time, precautions FROM herbs WHERE dosha = %s LIMIT %s" cursor.execute(q_herbs, (d_name, limit_val)) h_rows = cursor.fetchall() log_debug(f"Fetched {len(h_rows)} herbs for {d_name}") for h in h_rows: img = base_image_url + h['image_filename'] if h['image_filename'] else None final_herbs.append({ "name": h['name'], "description": h['description'] if h['description'] else "Ayurvedic Herb", "benefits": h['benefits'] if h['benefits'] else "Promotes general immunity.", "usage_dosage": h['usage_dosage'] if h['usage_dosage'] else "As directed.", "usage_preparation": h['usage_preparation'] if h['usage_preparation'] else "Powder.", "usage_time": h['usage_time'] if h['usage_time'] else "Morning.", "precautions": h['precautions'] if h['precautions'] else "None.", "image": img }) # E. Final Response (MATCHING LEGACY PHP STRUCTURE EXACTLY) # The Android App expects: recommendations -> {DominantDoshaName} -> {food, yoga...} # We constructed flat lists, now we nest them. nested_data = { "food": final_food, "yoga": final_yoga, "lifestyle": final_life, "herbs": final_herbs } response = { "status": "success", "dominant_dosha": dominant_dosha, "lowest_dosha": lowest_dosha, "recommendations": { dominant_dosha: nested_data # <--- CRITICAL NESTING }, "ai_mode": ML_AVAILABLE } log_debug("Success. Returning JSON (Legacy Format).") print(json.dumps(response)) except Exception as e: log_debug(f"CRITICAL ERROR: {e}") print(json.dumps({"status": "error", "message": str(e)})) finally: if conn and conn.is_connected(): conn.close() if __name__ == "__main__": if len(sys.argv) < 2: log_debug("Error: No User ID provided.") print(json.dumps({"status": "error", "message": "User ID required"})) else: get_recommendations(sys.argv[1])