remove backend

apps/core/tasks/__init__.py

@@ -0,0 +1,5 @@
+"""
+Core tasks package for ThrillWiki.
+
+This package contains all Celery tasks for the core application.
+"""

apps/core/tasks/trending.py

@@ -0,0 +1,607 @@
+"""
+Trending calculation tasks for ThrillWiki.
+
+This module contains Celery tasks for calculating and caching trending content.
+All tasks run asynchronously to avoid blocking the main application.
+"""
+
+import logging
+from datetime import datetime, timedelta
+from typing import Dict, List, Any
+from celery import shared_task
+from django.utils import timezone
+from django.core.cache import cache
+from django.contrib.contenttypes.models import ContentType
+from django.db.models import Q
+
+from apps.core.analytics import PageView
+from apps.parks.models import Park
+from apps.rides.models import Ride
+
+logger = logging.getLogger(__name__)
+
+
+@shared_task(bind=True, max_retries=3, default_retry_delay=60)
+def calculate_trending_content(
+    self, content_type: str = "all", limit: int = 50
+) -> Dict[str, Any]:
+    """
+    Calculate trending content using real analytics data.
+
+    This task runs periodically to update trending calculations based on:
+    - View growth rates
+    - Content ratings
+    - Recency factors
+    - Popularity metrics
+
+    Args:
+        content_type: 'parks', 'rides', or 'all'
+        limit: Maximum number of results to calculate
+
+    Returns:
+        Dict containing trending results and metadata
+    """
+    try:
+        logger.info(f"Starting trending calculation for {content_type}")
+
+        # Time windows for calculations
+        current_period_hours = 168  # 7 days
+        previous_period_hours = 336  # 14 days (for previous 7-day window comparison)
+
+        trending_items = []
+
+        if content_type in ["all", "parks"]:
+            park_items = _calculate_trending_parks(
+                current_period_hours,
+                previous_period_hours,
+                limit if content_type == "parks" else limit * 2,
+            )
+            trending_items.extend(park_items)
+
+        if content_type in ["all", "rides"]:
+            ride_items = _calculate_trending_rides(
+                current_period_hours,
+                previous_period_hours,
+                limit if content_type == "rides" else limit * 2,
+            )
+            trending_items.extend(ride_items)
+
+        # Sort by trending score and apply limit
+        trending_items.sort(key=lambda x: x.get("trending_score", 0), reverse=True)
+        trending_items = trending_items[:limit]
+
+        # Format results for API consumption
+        formatted_results = _format_trending_results(
+            trending_items, current_period_hours, previous_period_hours
+        )
+
+        # Cache results
+        cache_key = f"trending:calculated:{content_type}:{limit}"
+        cache.set(cache_key, formatted_results, 3600)  # Cache for 1 hour
+
+        logger.info(
+            f"Calculated {len(formatted_results)} trending items for {content_type}"
+        )
+
+        return {
+            "success": True,
+            "content_type": content_type,
+            "count": len(formatted_results),
+            "results": formatted_results,
+            "calculated_at": timezone.now().isoformat(),
+        }
+
+    except Exception as e:
+        logger.error(f"Error calculating trending content: {e}", exc_info=True)
+        # Retry the task
+        raise self.retry(exc=e)
+
+
+@shared_task(bind=True, max_retries=3, default_retry_delay=30)
+def calculate_new_content(
+    self, content_type: str = "all", days_back: int = 30, limit: int = 50
+) -> Dict[str, Any]:
+    """
+    Calculate new content based on opening dates and creation dates.
+
+    Args:
+        content_type: 'parks', 'rides', or 'all'
+        days_back: How many days to look back for new content
+        limit: Maximum number of results
+
+    Returns:
+        Dict containing new content results and metadata
+    """
+    try:
+        logger.info(f"Starting new content calculation for {content_type}")
+
+        cutoff_date = timezone.now() - timedelta(days=days_back)
+        new_items = []
+
+        if content_type in ["all", "parks"]:
+            parks = _get_new_parks(
+                cutoff_date, limit if content_type == "parks" else limit * 2
+            )
+            new_items.extend(parks)
+
+        if content_type in ["all", "rides"]:
+            rides = _get_new_rides(
+                cutoff_date, limit if content_type == "rides" else limit * 2
+            )
+            new_items.extend(rides)
+
+        # Sort by date added (most recent first) and apply limit
+        new_items.sort(key=lambda x: x.get("date_added", ""), reverse=True)
+        new_items = new_items[:limit]
+
+        # Format results for API consumption
+        formatted_results = _format_new_content_results(new_items)
+
+        # Cache results
+        cache_key = f"new_content:calculated:{content_type}:{days_back}:{limit}"
+        cache.set(cache_key, formatted_results, 1800)  # Cache for 30 minutes
+
+        logger.info(f"Calculated {len(formatted_results)} new items for {content_type}")
+
+        return {
+            "success": True,
+            "content_type": content_type,
+            "count": len(formatted_results),
+            "results": formatted_results,
+            "calculated_at": timezone.now().isoformat(),
+        }
+
+    except Exception as e:
+        logger.error(f"Error calculating new content: {e}", exc_info=True)
+        raise self.retry(exc=e)
+
+
+@shared_task(bind=True)
+def warm_trending_cache(self) -> Dict[str, Any]:
+    """
+    Warm the trending cache by pre-calculating common queries.
+
+    This task runs periodically to ensure fast API responses.
+    """
+    try:
+        logger.info("Starting trending cache warming")
+
+        # Common query combinations to pre-calculate
+        queries = [
+            {"content_type": "all", "limit": 20},
+            {"content_type": "parks", "limit": 10},
+            {"content_type": "rides", "limit": 10},
+            {"content_type": "all", "limit": 50},
+        ]
+
+        results = {}
+
+        for query in queries:
+            # Trigger trending calculation
+            calculate_trending_content.delay(**query)
+
+            # Trigger new content calculation
+            calculate_new_content.delay(**query)
+
+            results[f"trending_{query['content_type']}_{query['limit']}"] = "scheduled"
+            results[f"new_content_{query['content_type']}_{query['limit']}"] = (
+                "scheduled"
+            )
+
+        logger.info("Trending cache warming completed")
+
+        return {
+            "success": True,
+            "queries_scheduled": len(queries) * 2,
+            "results": results,
+            "warmed_at": timezone.now().isoformat(),
+        }
+
+    except Exception as e:
+        logger.error(f"Error warming trending cache: {e}", exc_info=True)
+        return {
+            "success": False,
+            "error": str(e),
+            "warmed_at": timezone.now().isoformat(),
+        }
+
+
+def _calculate_trending_parks(
+    current_period_hours: int, previous_period_hours: int, limit: int
+) -> List[Dict[str, Any]]:
+    """Calculate trending scores for parks using real data."""
+    parks = Park.objects.filter(status="OPERATING").select_related(
+        "location", "operator"
+    )
+
+    trending_parks = []
+
+    for park in parks:
+        try:
+            score = _calculate_content_score(
+                park, "park", current_period_hours, previous_period_hours
+            )
+            if score > 0:  # Only include items with positive trending scores
+                trending_parks.append(
+                    {
+                        "content_object": park,
+                        "content_type": "park",
+                        "trending_score": score,
+                        "id": park.id,
+                        "name": park.name,
+                        "slug": park.slug,
+                        "location": (
+                            park.formatted_location if hasattr(park, "location") else ""
+                        ),
+                        "category": "park",
+                        "rating": (
+                            float(park.average_rating) if park.average_rating else 0.0
+                        ),
+                    }
+                )
+        except Exception as e:
+            logger.warning(f"Error calculating score for park {park.id}: {e}")
+
+    return trending_parks
+
+
+def _calculate_trending_rides(
+    current_period_hours: int, previous_period_hours: int, limit: int
+) -> List[Dict[str, Any]]:
+    """Calculate trending scores for rides using real data."""
+    rides = Ride.objects.filter(status="OPERATING").select_related(
+        "park", "park__location"
+    )
+
+    trending_rides = []
+
+    for ride in rides:
+        try:
+            score = _calculate_content_score(
+                ride, "ride", current_period_hours, previous_period_hours
+            )
+            if score > 0:  # Only include items with positive trending scores
+                # Get location from park
+                location = ""
+                if ride.park and hasattr(ride.park, "location") and ride.park.location:
+                    location = ride.park.formatted_location
+
+                trending_rides.append(
+                    {
+                        "content_object": ride,
+                        "content_type": "ride",
+                        "trending_score": score,
+                        "id": ride.pk,
+                        "name": ride.name,
+                        "slug": ride.slug,
+                        "location": location,
+                        "category": "ride",
+                        "rating": (
+                            float(ride.average_rating) if ride.average_rating else 0.0
+                        ),
+                    }
+                )
+        except Exception as e:
+            logger.warning(f"Error calculating score for ride {ride.pk}: {e}")
+
+    return trending_rides
+
+
+def _calculate_content_score(
+    content_obj: Any,
+    content_type: str,
+    current_period_hours: int,
+    previous_period_hours: int,
+) -> float:
+    """
+    Calculate weighted trending score for content object using real analytics data.
+
+    Algorithm Components:
+    - View Growth Rate (40% weight): Recent view increase vs historical
+    - Rating Score (30% weight): Average user rating normalized
+    - Recency Factor (20% weight): How recently content was added/updated
+    - Popularity Boost (10% weight): Total view count normalization
+
+    Returns:
+        Float between 0.0 and 1.0 representing trending strength
+    """
+    try:
+        # Get content type for PageView queries
+        ct = ContentType.objects.get_for_model(content_obj)
+
+        # 1. View Growth Score (40% weight)
+        view_growth_score = _calculate_view_growth_score(
+            ct, content_obj.id, current_period_hours, previous_period_hours
+        )
+
+        # 2. Rating Score (30% weight)
+        rating_score = _calculate_rating_score(content_obj)
+
+        # 3. Recency Score (20% weight)
+        recency_score = _calculate_recency_score(content_obj)
+
+        # 4. Popularity Score (10% weight)
+        popularity_score = _calculate_popularity_score(
+            ct, content_obj.id, current_period_hours
+        )
+
+        # Calculate weighted final score
+        final_score = (
+            view_growth_score * 0.4
+            + rating_score * 0.3
+            + recency_score * 0.2
+            + popularity_score * 0.1
+        )
+
+        logger.debug(
+            f"{content_type} {content_obj.id}: "
+            f"growth={view_growth_score:.3f}, rating={rating_score:.3f}, "
+            f"recency={recency_score:.3f}, popularity={popularity_score:.3f}, "
+            f"final={final_score:.3f}"
+        )
+
+        return final_score
+
+    except Exception as e:
+        logger.error(
+            f"Error calculating score for {content_type} {content_obj.id}: {e}"
+        )
+        return 0.0
+
+
+def _calculate_view_growth_score(
+    content_type: ContentType,
+    object_id: int,
+    current_period_hours: int,
+    previous_period_hours: int,
+) -> float:
+    """Calculate normalized view growth score using real PageView data."""
+    try:
+        current_views, previous_views, growth_percentage = PageView.get_views_growth(
+            content_type,
+            object_id,
+            current_period_hours,
+            previous_period_hours,
+        )
+
+        if previous_views == 0:
+            # New content with views gets boost
+            return min(current_views / 100.0, 1.0) if current_views > 0 else 0.0
+
+        # Normalize growth percentage to 0-1 scale
+        # 100% growth = 0.5, 500% growth = 1.0
+        normalized_growth = (
+            min(growth_percentage / 500.0, 1.0) if growth_percentage > 0 else 0.0
+        )
+        return max(normalized_growth, 0.0)
+
+    except Exception as e:
+        logger.warning(f"Error calculating view growth: {e}")
+        return 0.0
+
+
+def _calculate_rating_score(content_obj: Any) -> float:
+    """Calculate normalized rating score."""
+    try:
+        rating = getattr(content_obj, "average_rating", None)
+        if rating is None or rating == 0:
+            return 0.3  # Neutral score for unrated content
+
+        # Normalize rating from 1-10 scale to 0-1 scale
+        # Rating of 5 = 0.4, Rating of 8 = 0.7, Rating of 10 = 1.0
+        return min(max((float(rating) - 1) / 9.0, 0.0), 1.0)
+
+    except Exception as e:
+        logger.warning(f"Error calculating rating score: {e}")
+        return 0.3
+
+
+def _calculate_recency_score(content_obj: Any) -> float:
+    """Calculate recency score based on when content was added/updated."""
+    try:
+        # Use opening_date for parks/rides, or created_at as fallback
+        date_added = getattr(content_obj, "opening_date", None)
+        if not date_added:
+            date_added = getattr(content_obj, "created_at", None)
+        if not date_added:
+            return 0.5  # Neutral score for unknown dates
+
+        # Handle both date and datetime objects
+        if hasattr(date_added, "date"):
+            date_added = date_added.date()
+
+        # Calculate days since added
+        today = timezone.now().date()
+        days_since_added = (today - date_added).days
+
+        # Recency score: newer content gets higher scores
+        # 0 days = 1.0, 30 days = 0.8, 365 days = 0.1, >365 days = 0.0
+        if days_since_added <= 0:
+            return 1.0
+        elif days_since_added <= 30:
+            return 1.0 - (days_since_added / 30.0) * 0.2  # 1.0 to 0.8
+        elif days_since_added <= 365:
+            return 0.8 - ((days_since_added - 30) / (365 - 30)) * 0.7  # 0.8 to 0.1
+        else:
+            return 0.0
+
+    except Exception as e:
+        logger.warning(f"Error calculating recency score: {e}")
+        return 0.5
+
+
+def _calculate_popularity_score(
+    content_type: ContentType, object_id: int, hours: int
+) -> float:
+    """Calculate popularity score based on total view count."""
+    try:
+        total_views = PageView.get_total_views_count(
+            content_type, object_id, hours=hours
+        )
+
+        # Normalize views to 0-1 scale
+        # 0 views = 0.0, 100 views = 0.5, 1000+ views = 1.0
+        if total_views == 0:
+            return 0.0
+        elif total_views <= 100:
+            return total_views / 200.0  # 0.0 to 0.5
+        else:
+            return min(0.5 + (total_views - 100) / 1800.0, 1.0)  # 0.5 to 1.0
+
+    except Exception as e:
+        logger.warning(f"Error calculating popularity score: {e}")
+        return 0.0
+
+
+def _get_new_parks(cutoff_date: datetime, limit: int) -> List[Dict[str, Any]]:
+    """Get recently added parks using real data."""
+    new_parks = (
+        Park.objects.filter(
+            Q(created_at__gte=cutoff_date) | Q(opening_date__gte=cutoff_date.date()),
+            status="OPERATING",
+        )
+        .select_related("location", "operator")
+        .order_by("-created_at", "-opening_date")[:limit]
+    )
+
+    results = []
+    for park in new_parks:
+        date_added = park.opening_date or park.created_at
+        if date_added:
+            if isinstance(date_added, datetime):
+                date_added = date_added.date()
+
+        opening_date = getattr(park, "opening_date", None)
+        if opening_date and isinstance(opening_date, datetime):
+            opening_date = opening_date.date()
+
+        results.append(
+            {
+                "content_object": park,
+                "content_type": "park",
+                "id": park.pk,
+                "name": park.name,
+                "slug": park.slug,
+                "park": park.name,  # For parks, park field is the park name itself
+                "category": "park",
+                "date_added": date_added.isoformat() if date_added else "",
+                "date_opened": opening_date.isoformat() if opening_date else "",
+            }
+        )
+
+    return results
+
+
+def _get_new_rides(cutoff_date: datetime, limit: int) -> List[Dict[str, Any]]:
+    """Get recently added rides using real data."""
+    new_rides = (
+        Ride.objects.filter(
+            Q(created_at__gte=cutoff_date) | Q(opening_date__gte=cutoff_date.date()),
+            status="OPERATING",
+        )
+        .select_related("park", "park__location")
+        .order_by("-created_at", "-opening_date")[:limit]
+    )
+
+    results = []
+    for ride in new_rides:
+        date_added = getattr(ride, "opening_date", None) or getattr(
+            ride, "created_at", None
+        )
+        if date_added:
+            if isinstance(date_added, datetime):
+                date_added = date_added.date()
+
+        opening_date = getattr(ride, "opening_date", None)
+        if opening_date and isinstance(opening_date, datetime):
+            opening_date = opening_date.date()
+
+        results.append(
+            {
+                "content_object": ride,
+                "content_type": "ride",
+                "id": ride.pk,
+                "name": ride.name,
+                "slug": ride.slug,
+                "park": ride.park.name if ride.park else "",
+                "category": "ride",
+                "date_added": date_added.isoformat() if date_added else "",
+                "date_opened": opening_date.isoformat() if opening_date else "",
+            }
+        )
+
+    return results
+
+
+def _format_trending_results(
+    trending_items: List[Dict[str, Any]],
+    current_period_hours: int,
+    previous_period_hours: int,
+) -> List[Dict[str, Any]]:
+    """Format trending results for frontend consumption."""
+    formatted_results = []
+
+    for rank, item in enumerate(trending_items, 1):
+        try:
+            # Get view change for display
+            content_obj = item["content_object"]
+            ct = ContentType.objects.get_for_model(content_obj)
+            current_views, previous_views, growth_percentage = (
+                PageView.get_views_growth(
+                    ct,
+                    content_obj.id,
+                    current_period_hours,
+                    previous_period_hours,
+                )
+            )
+
+            # Format exactly as frontend expects
+            formatted_item = {
+                "id": item["id"],
+                "name": item["name"],
+                "location": item["location"],
+                "category": item["category"],
+                "rating": item["rating"],
+                "rank": rank,
+                "views": current_views,
+                "views_change": (
+                    f"+{growth_percentage:.1f}%"
+                    if growth_percentage > 0
+                    else f"{growth_percentage:.1f}%"
+                ),
+                "slug": item["slug"],
+            }
+
+            formatted_results.append(formatted_item)
+
+        except Exception as e:
+            logger.warning(f"Error formatting trending item: {e}")
+
+    return formatted_results
+
+
+def _format_new_content_results(
+    new_items: List[Dict[str, Any]],
+) -> List[Dict[str, Any]]:
+    """Format new content results for frontend consumption."""
+    formatted_results = []
+
+    for item in new_items:
+        try:
+            # Format exactly as frontend expects
+            formatted_item = {
+                "id": item["id"],
+                "name": item["name"],
+                "park": item["park"],
+                "category": item["category"],
+                "date_added": item["date_added"],
+                "date_opened": item["date_opened"],
+                "slug": item["slug"],
+            }
+
+            formatted_results.append(formatted_item)
+
+        except Exception as e:
+            logger.warning(f"Error formatting new content item: {e}")
+
+    return formatted_results