updated dashboard

src/exchanges/deutsche_boerse.py

@@ -34,6 +34,10 @@ class DeutscheBoerseBase(BaseExchange):
             response.raise_for_status()
             
             files = []
+            html_text = response.text
+            
+            # Debug: Response-Länge
+            print(f"[{self.name}] Response length: {len(html_text)} chars")
             
             # Primär: Regex-basierte Extraktion (zuverlässiger)
             # Pattern: PREFIX-posttrade-YYYY-MM-DDTHH_MM.json.gz
@@ -47,13 +51,16 @@ class DeutscheBoerseBase(BaseExchange):
                 # Generisches Pattern
                 pattern = r'[A-Z]{4}-posttrade-\d{4}-\d{2}-\d{2}T\d{2}_\d{2}\.json\.gz'
             
-            matches = re.findall(pattern, response.text)
+            matches = re.findall(pattern, html_text)
             files = list(set(matches))
             
             # Sekundär: BeautifulSoup für Links (falls Regex nichts findet)
             if not files:
-                soup = BeautifulSoup(response.text, 'html.parser')
-                for link in soup.find_all('a'):
+                soup = BeautifulSoup(html_text, 'html.parser')
+                all_links = soup.find_all('a')
+                print(f"[{self.name}] Found {len(all_links)} total links on page")
+                
+                for link in all_links:
                     href = link.get('href', '')
                     text = link.get_text(strip=True)
                     
@@ -65,6 +72,15 @@ class DeutscheBoerseBase(BaseExchange):
                     elif text and 'posttrade' in text.lower() and '.json.gz' in text.lower():
                         files.append(text)
             
+            # Tertiär: Suche nach jedem "posttrade" im HTML und extrahiere Dateinamen
+            if not files:
+                # Allgemeineres Pattern für beliebige Dateinamen mit "posttrade"
+                general_pattern = r'[\w-]*posttrade[\w-]*\d{4}[-_]\d{2}[-_]\d{2}[T_]\d{2}[_:]\d{2}\.json\.gz'
+                matches = re.findall(general_pattern, html_text, re.IGNORECASE)
+                files = list(set(matches))
+                if files:
+                    print(f"[{self.name}] Found {len(files)} files via general pattern")
+            
             print(f"[{self.name}] Found {len(files)} files via regex/soup")
             return files
         except Exception as e:
@@ -233,6 +249,34 @@ class DeutscheBoerseBase(BaseExchange):
             print(f"Error parsing record: {e}")
             return None
     
+    def _generate_expected_files(self, target_date: datetime.date) -> List[str]:
+        """
+        Generiert erwartete Dateinamen basierend auf dem bekannten Format.
+        Format: PREFIX-posttrade-YYYY-MM-DDTHH_MM.json.gz
+        """
+        import re
+        files = []
+        
+        # Extrahiere Prefix aus base_url (z.B. DETR, DFRA, DGAT)
+        prefix_match = re.search(r'/([A-Z]{4})-posttrade', self.base_url)
+        prefix = prefix_match.group(1) if prefix_match else 'DETR'
+        
+        date_str = target_date.strftime('%Y-%m-%d')
+        
+        # Generiere für alle Stunden des Handelstages (07:00 - 22:00 UTC, alle Minuten)
+        for hour in range(7, 23):
+            for minute in range(0, 60):
+                files.append(f"{prefix}-posttrade-{date_str}T{hour:02d}_{minute:02d}.json.gz")
+        
+        # Auch frühe Dateien vom Folgetag (nach Mitternacht UTC)
+        next_date = target_date + timedelta(days=1)
+        next_date_str = next_date.strftime('%Y-%m-%d')
+        for hour in range(0, 3):
+            for minute in range(0, 60):
+                files.append(f"{prefix}-posttrade-{next_date_str}T{hour:02d}_{minute:02d}.json.gz")
+        
+        return files
+
     def fetch_latest_trades(self, include_yesterday: bool = True, since_date: datetime = None) -> List[Trade]:
         """
         Holt alle Trades vom Vortag (oder seit since_date).
@@ -248,7 +292,7 @@ class DeutscheBoerseBase(BaseExchange):
         
         print(f"[{self.name}] Fetching trades for date: {target_date}")
         
-        # Dateiliste holen
+        # Erst versuchen, Dateiliste von der Seite zu holen
         files = self._get_file_list()
         print(f"[{self.name}] Found {len(files)} total files")
         
@@ -256,11 +300,24 @@ class DeutscheBoerseBase(BaseExchange):
         target_files = self._filter_files_for_date(files, target_date)
         print(f"[{self.name}] {len(target_files)} files match target date")
         
+        # Falls keine Dateien von der Seite gefunden, generiere erwartete Dateinamen
+        if not target_files:
+            print(f"[{self.name}] No files from page, trying generated filenames...")
+            target_files = self._generate_expected_files(target_date)
+            print(f"[{self.name}] Trying {len(target_files)} potential files")
+        
         # Alle passenden Dateien herunterladen und parsen
+        successful = 0
         for file in target_files:
             trades = self._download_and_parse_file(file)
-            all_trades.extend(trades)
-            print(f"[{self.name}] Parsed {len(trades)} trades from {file}")
+            if trades:
+                all_trades.extend(trades)
+                successful += 1
+                if successful <= 5:
+                    print(f"[{self.name}] Parsed {len(trades)} trades from {file}")
+        
+        if successful > 5:
+            print(f"[{self.name}] ... and {successful - 5} more files")
         
         print(f"[{self.name}] Total trades fetched: {len(all_trades)}")
         return all_trades