fix: Auction end_time timezone - add UTC suffix

The frontend was calculating wrong time remaining because
end_time was sent without timezone suffix (Z).

JavaScript's new Date() interprets "2025-12-20T20:49:20" as
local time, but the server stores it as UTC.

Fix: Add json_encoders to Pydantic models that append "Z"
to all datetime fields, marking them as UTC.

Affected models:
- AuctionListing
- AuctionSearchResponse
- MarketFeedItem
- MarketFeedResponse
- ScrapeStatus

.gitea/workflows/deploy.yml

@@ -62,11 +62,21 @@ jobs:
           docker stop pounce-backend 2>/dev/null || true
           docker rm pounce-backend 2>/dev/null || true
           
+          # Ensure persistent directories exist
+          sudo mkdir -p /data/pounce/zones/czds /data/pounce/zones/switch /data/pounce/logs
+          sudo chmod -R 755 /data/pounce
+          
           # Run new container with secrets from environment
           docker run -d \
             --name pounce-backend \
             --network n0488s44osgoow4wgo04ogg0 \
             --restart unless-stopped \
+            -v /data/pounce/zones/czds:/data/czds \
+            -v /data/pounce/zones/switch:/data/switch \
+            -v /data/pounce/logs:/data/logs \
+            -e CZDS_DATA_DIR="/data/czds" \
+            -e SWITCH_DATA_DIR="/data/switch" \
+            -e ZONE_RETENTION_DAYS="3" \
             -e DATABASE_URL="${DATABASE_URL}" \
             -e SECRET_KEY="${SECRET_KEY}" \
             -e JWT_SECRET="${SECRET_KEY}" \

backend/app/api/auctions.py

@@ -78,6 +78,10 @@ class AuctionListing(BaseModel):
     
     class Config:
         from_attributes = True
+        # Serialize datetimes as ISO format with UTC timezone suffix
+        json_encoders = {
+            datetime: lambda v: v.isoformat() + "Z" if v else None
+        }
 
 
 class AuctionSearchResponse(BaseModel):
@@ -92,6 +96,11 @@ class AuctionSearchResponse(BaseModel):
         "$50 × Length × TLD × Keyword × Brand factors. "
         "See /portfolio/valuation/{domain} for detailed breakdown."
     )
+    
+    class Config:
+        json_encoders = {
+            datetime: lambda v: v.isoformat() + "Z" if v else None
+        }
 
 
 class PlatformStats(BaseModel):
@@ -108,6 +117,11 @@ class ScrapeStatus(BaseModel):
     total_auctions: int
     platforms: List[str]
     next_scrape: Optional[datetime]
+    
+    class Config:
+        json_encoders = {
+            datetime: lambda v: v.isoformat() + "Z" if v else None
+        }
 
 
 class MarketFeedItem(BaseModel):
@@ -146,6 +160,9 @@ class MarketFeedItem(BaseModel):
     
     class Config:
         from_attributes = True
+        json_encoders = {
+            datetime: lambda v: v.isoformat() + "Z" if v else None
+        }
 
 
 class MarketFeedResponse(BaseModel):
@@ -157,6 +174,11 @@ class MarketFeedResponse(BaseModel):
     sources: List[str]
     last_updated: datetime
     filters_applied: dict = {}
+    
+    class Config:
+        json_encoders = {
+            datetime: lambda v: v.isoformat() + "Z" if v else None
+        }
 
 
 # ============== Helper Functions ==============

backend/app/config.py

@@ -129,7 +129,13 @@ class Settings(BaseSettings):
     # Register at: https://czds.icann.org/
     czds_username: str = ""
     czds_password: str = ""
-    czds_data_dir: str = "/tmp/pounce_czds"
+    czds_data_dir: str = "/data/czds"  # Persistent storage
+    
+    # Switch.ch Zone Files (.ch, .li)
+    switch_data_dir: str = "/data/switch"  # Persistent storage
+    
+    # Zone File Retention (days to keep historical snapshots)
+    zone_retention_days: int = 3
     
     class Config:
         env_file = ".env"

backend/app/services/zone_file.py

@@ -49,7 +49,9 @@ class ZoneFileService:
     """Service for fetching and analyzing zone files"""
     
     def __init__(self, data_dir: Optional[Path] = None):
-        self.data_dir = data_dir or Path("/tmp/pounce_zones")
+        from app.config import get_settings
+        settings = get_settings()
+        self.data_dir = data_dir or Path(settings.switch_data_dir)
         self.data_dir.mkdir(parents=True, exist_ok=True)
     
     def _get_key_file_path(self, tld: str) -> Path:

backend/app/services/zone_retention.py

@@ -0,0 +1,230 @@
+"""
+Zone File Retention Management
+==============================
+Manages historical zone file snapshots with configurable retention period.
+Default: 3 days of history for reliable drop detection.
+
+Features:
+- Daily snapshots with timestamps
+- Automatic cleanup of old snapshots
+- Reliable diff calculation across multiple days
+"""
+
+import logging
+import shutil
+from datetime import datetime, timedelta
+from pathlib import Path
+from typing import Optional
+
+from app.config import get_settings
+
+logger = logging.getLogger(__name__)
+settings = get_settings()
+
+
+class ZoneRetentionManager:
+    """
+    Manages zone file snapshots with retention policy.
+    
+    Directory structure:
+        /data/czds/
+            xyz_domains.txt          <- current/latest
+            xyz_domains_2024-01-15.txt  <- daily snapshot
+            xyz_domains_2024-01-14.txt
+            xyz_domains_2024-01-13.txt
+    """
+    
+    def __init__(self, data_dir: Optional[Path] = None, retention_days: int = 3):
+        self.data_dir = data_dir or Path(settings.czds_data_dir)
+        self.retention_days = retention_days or settings.zone_retention_days
+        self.data_dir.mkdir(parents=True, exist_ok=True)
+    
+    def get_snapshot_path(self, tld: str, date: datetime) -> Path:
+        """Get path for a dated snapshot."""
+        date_str = date.strftime("%Y-%m-%d")
+        return self.data_dir / f"{tld}_domains_{date_str}.txt"
+    
+    def get_current_path(self, tld: str) -> Path:
+        """Get path for current (latest) snapshot."""
+        return self.data_dir / f"{tld}_domains.txt"
+    
+    def save_snapshot(self, tld: str, domains: set[str], date: Optional[datetime] = None):
+        """
+        Save a domain snapshot with date suffix and update current.
+        
+        Args:
+            tld: The TLD (e.g., 'xyz', 'ch')
+            domains: Set of domain names
+            date: Optional date for snapshot (defaults to today)
+        """
+        date = date or datetime.utcnow()
+        
+        # Save dated snapshot
+        snapshot_path = self.get_snapshot_path(tld, date)
+        content = "\n".join(sorted(domains))
+        snapshot_path.write_text(content)
+        
+        # Also update current pointer
+        current_path = self.get_current_path(tld)
+        current_path.write_text(content)
+        
+        logger.info(f"Saved .{tld} snapshot: {len(domains):,} domains -> {snapshot_path.name}")
+    
+    def load_snapshot(self, tld: str, date: Optional[datetime] = None) -> Optional[set[str]]:
+        """
+        Load a snapshot from a specific date.
+        
+        Args:
+            tld: The TLD
+            date: Date to load (None = current/latest)
+        
+        Returns:
+            Set of domain names or None if not found
+        """
+        if date:
+            path = self.get_snapshot_path(tld, date)
+        else:
+            path = self.get_current_path(tld)
+        
+        if not path.exists():
+            return None
+        
+        try:
+            content = path.read_text()
+            return set(line.strip() for line in content.splitlines() if line.strip())
+        except Exception as e:
+            logger.warning(f"Failed to load snapshot {path.name}: {e}")
+            return None
+    
+    def get_previous_snapshot(self, tld: str, days_ago: int = 1) -> Optional[set[str]]:
+        """
+        Load snapshot from N days ago.
+        
+        Args:
+            tld: The TLD
+            days_ago: How many days back to look
+        
+        Returns:
+            Set of domain names or None
+        """
+        target_date = datetime.utcnow() - timedelta(days=days_ago)
+        return self.load_snapshot(tld, target_date)
+    
+    def cleanup_old_snapshots(self, tld: Optional[str] = None) -> int:
+        """
+        Remove snapshots older than retention period.
+        
+        Args:
+            tld: Optional TLD to clean (None = all TLDs)
+        
+        Returns:
+            Number of files deleted
+        """
+        cutoff_date = datetime.utcnow() - timedelta(days=self.retention_days)
+        deleted = 0
+        
+        # Pattern: *_domains_YYYY-MM-DD.txt
+        pattern = f"{tld}_domains_*.txt" if tld else "*_domains_*.txt"
+        
+        for file_path in self.data_dir.glob(pattern):
+            # Skip current files (no date suffix)
+            name = file_path.stem
+            if not any(c.isdigit() for c in name):
+                continue
+            
+            # Extract date from filename
+            try:
+                # Get the date part (last 10 chars: YYYY-MM-DD)
+                date_str = name[-10:]
+                file_date = datetime.strptime(date_str, "%Y-%m-%d")
+                
+                if file_date < cutoff_date:
+                    file_path.unlink()
+                    deleted += 1
+                    logger.info(f"Deleted old snapshot: {file_path.name}")
+            except (ValueError, IndexError):
+                # Not a dated snapshot, skip
+                continue
+        
+        if deleted > 0:
+            logger.info(f"Cleaned up {deleted} old zone file snapshots")
+        
+        return deleted
+    
+    def get_available_snapshots(self, tld: str) -> list[datetime]:
+        """
+        List all available snapshot dates for a TLD.
+        
+        Args:
+            tld: The TLD
+        
+        Returns:
+            List of dates (sorted, newest first)
+        """
+        dates = []
+        pattern = f"{tld}_domains_*.txt"
+        
+        for file_path in self.data_dir.glob(pattern):
+            name = file_path.stem
+            try:
+                date_str = name[-10:]
+                file_date = datetime.strptime(date_str, "%Y-%m-%d")
+                dates.append(file_date)
+            except (ValueError, IndexError):
+                continue
+        
+        return sorted(dates, reverse=True)
+    
+    def get_storage_stats(self) -> dict:
+        """Get storage statistics for zone files."""
+        stats = {
+            "total_files": 0,
+            "total_size_mb": 0.0,
+            "tlds": {},
+        }
+        
+        for file_path in self.data_dir.glob("*_domains*.txt"):
+            stats["total_files"] += 1
+            size_mb = file_path.stat().st_size / (1024 * 1024)
+            stats["total_size_mb"] += size_mb
+            
+            # Extract TLD
+            name = file_path.stem
+            tld = name.split("_")[0]
+            if tld not in stats["tlds"]:
+                stats["tlds"][tld] = {"files": 0, "size_mb": 0.0}
+            stats["tlds"][tld]["files"] += 1
+            stats["tlds"][tld]["size_mb"] += size_mb
+        
+        return stats
+
+
+def migrate_existing_snapshots():
+    """
+    Migrate existing zone files to dated snapshot format.
+    Call this once during deployment.
+    """
+    manager = ZoneRetentionManager()
+    today = datetime.utcnow()
+    migrated = 0
+    
+    for data_dir in [Path(settings.czds_data_dir), Path(settings.switch_data_dir)]:
+        if not data_dir.exists():
+            continue
+        
+        for file_path in data_dir.glob("*_domains.txt"):
+            name = file_path.stem
+            # Skip if already has date
+            if any(c.isdigit() for c in name[-10:]):
+                continue
+            
+            tld = name.replace("_domains", "")
+            
+            # Create dated copy
+            dated_path = data_dir / f"{tld}_domains_{today.strftime('%Y-%m-%d')}.txt"
+            if not dated_path.exists():
+                shutil.copy(file_path, dated_path)
+                migrated += 1
+                logger.info(f"Migrated {file_path.name} -> {dated_path.name}")
+    
+    return migrated

backend/scripts/sync_all_zones.py

@@ -58,7 +58,7 @@ SWITCH_CONFIG = {
     "li": {
         "server": "zonedata.switch.ch", 
         "key_name": "tsig-zonedata-li-public-21-01.",
-        "key_secret": "t8GgeCn+fhPaj+cRy/lakQPb6M45xz/NZwmcp4iqbBxKFCCH0/k3xNGe6sf3ObmoaKDBedge/La4cpPfLqtFkw=="
+        "key_secret": "t8GgeCn+fhPaj+cRy1epox2Vj4hZ45ax6v3rQCkkfIQNg5fsxuU23QM5mzz+BxJ4kgF/jiQyBDBvL+XWPE6oCQ=="
     }
 }