scripts/recalculate_analytics.py

import os
import sys
from pathlib import Path
from datetime import datetime, timezone, timedelta

# Add the server root to the Python path
server_root = Path(__file__).resolve().parent.parent
sys.path.append(str(server_root))

from firebase_admin import firestore
from src.db.firebase import get_firebase_db
from src.utils.logger import setup_logger

logger = setup_logger(__name__)

def recalculate_analytics():
    """
    Recalculates all KPI fields for every user by iterating through their notes.
    """
    db = get_firebase_db()
    if not db:
        logger.error("Database not initialized")
        return

    logger.info("Starting comprehensive analytics recalculation...")

    users_ref = db.collection('users')
    notes_ref = db.collection('notes')
    analytics_ref = db.collection('analytics')

    try:
        users = users_ref.stream()
        migrated_count = 0

        now = datetime.now(timezone.utc)
        week_ago = now - timedelta(days=7)
        current_month = now.month
        current_year = now.year

        for user_doc in users:
            user_id = user_doc.id
            logger.info(f"Processing user: {user_id}")

            # Fetch all notes for this user
            user_notes = list(notes_ref.where('userId', '==', user_id).stream())
            
            # --- Initialize KPI Counters ---
            notes_count = len(user_notes)
            category_count = {}
            favorite_notes_count = 0
            total_key_points = 0
            total_video_duration_seconds = 0
            this_week_videos = 0
            this_month_saved_hours = 0.0
            
            # For streak calculation
            note_dates = set()

            for note_doc in user_notes:
                data = note_doc.to_dict()
                
                # Categories
                categories = []
                if 'category' in data:
                    cat_data = data['category']
                    if isinstance(cat_data, list):
                        categories = cat_data
                    elif isinstance(cat_data, str):
                        categories = [cat_data]
                elif 'video_categories' in data and isinstance(data['video_categories'], list):
                    categories = data['video_categories']
                elif 'categories' in data and isinstance(data['categories'], list):
                    categories = data['categories']
                
                if 'Uncategorized' in categories:
                    categories.remove('Uncategorized')
                if not categories:
                    categories.append('Technology & AI')

                for cat in categories:
                    category_count[cat] = category_count.get(cat, 0) + 1

                # Favorites
                if data.get('isFavorite', False):
                    favorite_notes_count += 1
                
                # Key Points
                key_points = data.get('keyPoints', [])
                total_key_points += len(key_points)

                # Duration
                video_duration = data.get('videoDuration', data.get('video_duration', 0))
                total_video_duration_seconds += video_duration

                # Dates (for streak, weekly, monthly stats)
                created_at = data.get('createdAt')
                if created_at:
                    dt = created_at
                    # Some data might have python datetimes if created directly
                    if hasattr(created_at, 'timestamp'): 
                        # Handle Google Cloud Timestamp vs Python datetime
                        dt_val = created_at
                    
                    # Store date string for streak calculation (YYYY-MM-DD)
                    note_dates.add(dt.strftime('%Y-%m-%d'))
                    
                    # This week videos
                    if dt >= week_ago:
                        this_week_videos += 1
                        
                    # This month saved hours (assuming videoDuration is the saved time roughly)
                    if dt.month == current_month and dt.year == current_year:
                        this_month_saved_hours += video_duration / 3600.0

            # --- Finalize Aggregations ---
            total_minutes = total_video_duration_seconds // 60
            total_saved_hours = total_video_duration_seconds / 3600.0

            # Find favorite category
            favorite_category = 'None'
            if category_count:
                favorite_category = max(category_count.items(), key=lambda x: x[1])[0]

            # Calculate current streak
            current_streak = 0
            check_date = now.date()
            # If they haven't posted today, check if they posted yesterday (streak is still alive)
            if check_date.strftime('%Y-%m-%d') not in note_dates:
                check_date = check_date - timedelta(days=1)
                
            while check_date.strftime('%Y-%m-%d') in note_dates:
                current_streak += 1
                check_date -= timedelta(days=1)

            # Update analytics document
            analytics_data = {
                'userId': user_id,
                'notesCount': notes_count,
                'categoryCount': category_count,
                'favoriteCategory': favorite_category,
                'favoriteNotesCount': favorite_notes_count,
                'totalKeyPoints': total_key_points,
                'totalMinutes': total_minutes,
                'totalSavedHours': round(total_saved_hours, 2),
                'thisWeekVideos': this_week_videos,
                'thisMonthSavedHours': round(this_month_saved_hours, 2),
                'currentStreak': current_streak,
                'lastUpdated': firestore.SERVER_TIMESTAMP,
            }

            analytics_ref.document(user_id).set(analytics_data, merge=True)
            logger.info(f"Updated analytics for {user_id}: {notes_count} notes, {current_streak} streak.")
            
            migrated_count += 1

        logger.info(f"Recalculation completed. Processed {migrated_count} users.")

    except Exception as e:
        logger.error(f"Error during recalculation: {e}", exc_info=True)

if __name__ == "__main__":
    recalculate_analytics()