auf Webseiten Modell umgestellt

backend/app/providers/barclays_arena.py

@@ -1,4 +1,5 @@
 from datetime import datetime
+import re
 from urllib.parse import urljoin
 
 import requests
@@ -10,7 +11,7 @@ from app.providers.utils import normalize_search_text
 
 class BarclaysArenaProvider:
     source_name = "barclays_arena"
-    events_url = "https://www.barclays-arena.de/events"
+    events_url = "https://www.barclays-arena.de/events/search"
 
     def search_events(
         self,
@@ -29,73 +30,104 @@ class BarclaysArenaProvider:
         normalized_term = normalize_search_text(term)
         results: list[dict] = []
 
-        headings = soup.find_all("h3")
-        for heading in headings:
+        for heading in soup.find_all("h3"):
             title = heading.get_text(" ", strip=True)
             if not title:
                 continue
 
-            subtitle = ""
-            subtitle_el = heading.find_next("h4")
-            if subtitle_el:
-                subtitle = subtitle_el.get_text(" ", strip=True)
+            subtitle_el = heading.find_next_sibling("h4")
+            subtitle = subtitle_el.get_text(" ", strip=True) if subtitle_el else ""
 
+            # Keep matching local to the actual heading/subtitle pair. Wider
+            # parent containers often contain several event cards.
             haystack = normalize_search_text(f"{title} {subtitle}")
             if normalized_term not in haystack:
                 continue
 
-            date_text = self._find_previous_date_text(heading)
-            event_date = self._parse_german_date(date_text)
-
-            link = heading.find_previous("a", href=True)
-            if link is None:
+            detail_link = self._find_card_link(heading)
+            if detail_link is None:
                 continue
 
+            date_text = self._find_card_date_text(heading)
+            event_date = self._parse_german_date(date_text)
+            href = detail_link["href"]
+
             results.append(
                 {
-                    "external_id": link["href"],
+                    "external_id": href,
                     "title": title,
                     "matched_term": term,
                     "venue_name": "Barclays Arena",
                     "city": "Hamburg",
                     "country_code": "DE",
                     "event_date": event_date,
-                    "ticket_url": urljoin(self.events_url, link["href"]),
+                    "ticket_url": urljoin(self.events_url, href),
                     "image_url": None,
                     "raw_payload": {
                         "title": title,
                         "subtitle": subtitle,
                         "date_text": date_text,
-                        "href": link["href"],
+                        "href": href,
                     },
                 }
             )
 
+        unique_results: dict[str, dict] = {}
+        for result in results:
+            unique_results[result["external_id"]] = result
+
         self.last_status = "ok"
         self.last_message = (
-            f"Barclays Arena returned {len(results)} matched events for term '{term}'."
+            f"Barclays Arena returned {len(unique_results)} matched events for term '{term}'."
         )
-        return results
+        return list(unique_results.values())
 
-    def _find_previous_date_text(self, heading) -> str | None:
-        current = heading.previous_sibling
-        while current is not None:
+    def _find_card_link(self, heading):
+        link = heading.find_parent("a", href=re.compile(r"/events/"))
+        if link is not None:
+            return link
+
+        current = heading
+        for _ in range(5):
+            current = current.parent
+            if current is None:
+                return None
+            link = current.find("a", href=re.compile(r"/events/"))
+            if link is not None and heading in link.find_all("h3"):
+                return link
+        return None
+
+    def _find_card_date_text(self, heading) -> str | None:
+        current = heading
+        for _ in range(6):
+            current = current.previous_element
+            if current is None:
+                return None
             text = getattr(current, "get_text", lambda *args, **kwargs: str(current))(
                 " ", strip=True
             )
-            if text and "|" in text:
-                return text
-            current = current.previous_sibling
+            date_text = self._extract_date_text(text)
+            if date_text:
+                return date_text
+        return None
+
+    def _extract_date_text(self, text: str) -> str | None:
+        match = re.search(
+            r"(Montag|Dienstag|Mittwoch|Donnerstag|Freitag|Samstag|Sonntag)\s*\|\s*(\d{2}\.\d{2}\.\d{4})",
+            text,
+        )
+        if match:
+            return match.group(2)
+
+        match = re.search(r"\b(\d{2}\.\d{2}\.\d{4})\b", text)
+        if match:
+            return match.group(1)
         return None
 
     def _parse_german_date(self, value: str | None) -> datetime | None:
         if not value:
             return None
-        parts = [part.strip() for part in value.split("|")]
-        if len(parts) < 2:
-            return None
         try:
-            return datetime.strptime(parts[1], "%d.%m.%Y")
+            return datetime.strptime(value, "%d.%m.%Y")
         except ValueError:
             return None
-