fix: Momondo Consent+URL, Wego URL-Format, Traveloka USD→EUR Parsing

Co-authored-by: Cursor <cursoragent@cursor.com>

node/src/worker.py

@@ -115,7 +115,7 @@ def _booking_url_momondo(von, nach, abflug, rueck, kc, bags=1,
     if airline:
         filters.append(f"airlines%3D{airline}")
     fs = ("&fs=" + "%3B".join(filters)) if filters else ""
-    base = f"https://www.momondo.de/flightsearch/{von}-{nach}/{abflug}"
+    base = f"https://www.momondo.de/flight-search/{von}-{nach}/{abflug}"
     if rueck:
         return f"{base}/{rueck}?sort=price_a&cabin={kc}&currency=EUR{fs}"
     return f"{base}?sort=price_a&cabin={kc}&currency=EUR{fs}"
@@ -726,11 +726,25 @@ def scrape_momondo(von, nach, tage=30, aufenthalt_tage=60,
 
     with SB(uc=True, headless=True, chromium_arg="--no-sandbox --disable-dev-shm-usage") as sb:
         sb.open(booking_url)
-        sb.sleep(15)
+        sb.sleep(8)
 
+        # Momondo Cookie-Consent wegklicken
+        for sel in ['button[class*="accept"]', '.RxNS-button-content',
+                    '#onetrust-accept-btn-handler', 'button[title*="akzeptieren"]',
+                    'button[title*="Alle akzeptieren"]', '.evidon-banner-acceptbutton']:
+            try:
+                sb.find_element(sel, timeout=2).click()
+                print(f"[MO] Consent geklickt: {sel}")
+                sb.sleep(3)
+                break
+            except Exception:
+                pass
+
+        # Nach Consent: Seite muss neu laden / Ergebnisse warten
+        sb.sleep(12)
         title = sb.get_title()
         body = sb.get_text("body")
-        print(f"[MO] Title: {title[:80]}")
+        print(f"[MO] Title: {title[:80]} | Body: {len(body)} chars")
 
         for sel in ['.price-text', '.f8F1-price-text', 'div[class*="price"] span',
                     'span[class*="price"]', '.Iqt3', 'div.nrc6-price', '.price',
@@ -789,18 +803,18 @@ def scrape_wego(von, nach, tage=30, aufenthalt_tage=60,
                    "business": "business", "first": "first"}
     kc = KABINE_WEGO.get(kabine, "premiumEconomy")
 
+    stadtname_wego = {"FRA": "frankfurt", "KTI": "phnom-penh", "HAN": "hanoi",
+                      "BKK": "bangkok", "SGN": "ho-chi-minh-city", "HKG": "hong-kong"}
+    von_slug  = stadtname_wego.get(von, von.lower())
+    nach_slug = stadtname_wego.get(nach, nach.lower())
     if rueck:
-        booking_url = (f"https://www.wego.com/flights/searches/new?"
+        booking_url = (f"https://www.wego.com/flights/{von.lower()}/{nach.lower()}"
-                       f"origin={von.lower()}&destination={nach.lower()}"
+                       f"/{abflug}/{rueck}"
-                       f"&outbound_date={abflug}&inbound_date={rueck}"
+                       f"?cabin_class={kc}&adults_count=1&sort=price&currency_code=EUR")
-                       f"&cabin={kc}&adults=1&children=0&infants=0"
-                       f"&currency=EUR&sort=price")
     else:
-        booking_url = (f"https://www.wego.com/flights/searches/new?"
+        booking_url = (f"https://www.wego.com/flights/{von.lower()}/{nach.lower()}"
-                       f"origin={von.lower()}&destination={nach.lower()}"
+                       f"/{abflug}"
-                       f"&outbound_date={abflug}"
+                       f"?cabin_class={kc}&adults_count=1&sort=price&currency_code=EUR")
-                       f"&cabin={kc}&adults=1&children=0&infants=0"
-                       f"&currency=EUR&sort=price")
 
     print(f"[WG] URL: {booking_url}")
     results = []
@@ -843,12 +857,30 @@ def scrape_wego(von, nach, tage=30, aufenthalt_tage=60,
     return results[:10], screenshot_b64
 
 
+def _parse_preis_usd(text):
+    """Parst USD-Preise aus Text wie 'USD 1,388.60' und wandelt grob in EUR um."""
+    if not text:
+        return None
+    # USD-Format: 1,388.60 (Komma als Tausender, Punkt als Dezimal)
+    for p in [r'USD\s?([\d,]+\.?\d*)', r'\$\s?([\d,]+\.?\d*)']:
+        m = re.search(p, text)
+        if m:
+            try:
+                v = float(m.group(1).replace(',', ''))
+                eur = round(v * 0.92, 2)  # grobe USD→EUR Umrechnung
+                if 200 < eur < 15000:
+                    return eur
+            except ValueError:
+                pass
+    return None
+
+
 def scrape_traveloka(von, nach, tage=30, aufenthalt_tage=60,
                      trip_type="roundtrip", kabine="premium_economy",
                      gepaeck="1koffer+handgepaeck", airline_filter="",
                      layover_min=120, layover_max=300,
                      max_flugzeit_h=22, max_stops=2):
-    """Traveloka — größte Reiseplattform Südostasiens."""
+    """Traveloka — größte Reiseplattform Südostasiens. Preise in USD, werden in EUR umgerechnet."""
     abflug = (datetime.now() + timedelta(days=tage)).strftime("%d-%m-%Y")
     rueck  = (datetime.now() + timedelta(days=tage + aufenthalt_tage)).strftime("%d-%m-%Y") \
              if trip_type == "roundtrip" else ""
@@ -856,7 +888,6 @@ def scrape_traveloka(von, nach, tage=30, aufenthalt_tage=60,
     rueck_iso  = (datetime.now() + timedelta(days=tage + aufenthalt_tage)).strftime("%Y-%m-%d") \
                  if trip_type == "roundtrip" else ""
 
-    # Traveloka URL-Parameter
     KABINE_TV = {"economy": "ECONOMY", "premium_economy": "PREMIUM_ECONOMY",
                  "business": "BUSINESS", "first": "FIRST_CLASS"}
     kc = KABINE_TV.get(kabine, "PREMIUM_ECONOMY")
@@ -882,40 +913,22 @@ def scrape_traveloka(von, nach, tage=30, aufenthalt_tage=60,
         body = sb.get_text("body")
         print(f"[TV] Title: {title[:80]} | Body: {len(body)} chars")
 
-        # Cookie/Consent Banner klicken
+        # Preise aus dem Body-Text extrahieren (USD → EUR)
-        for sel in ['button[data-testid*="accept"]', 'button[id*="accept"]',
+        seen = set()
-                    '#onetrust-accept-btn-handler', 'button[class*="accept"]']:
+        for m in re.finditer(r'USD\s?([\d,]+\.?\d*)', body):
             try:
-                sb.find_element(sel, timeout=2).click()
+                usd = float(m.group(1).replace(',', ''))
-                sb.sleep(2)
+                eur = round(usd * 0.92)
-                break
+                if 400 < eur < 12000 and eur not in seen:
-            except Exception:
+                    seen.add(eur)
+                    results.append({"scanner": "traveloka", "preis": eur,
+                                    "waehrung": "EUR", "airline": "",
+                                    "abflug": abflug_iso, "ankunft": rueck_iso,
+                                    "booking_url": booking_url})
+            except ValueError:
                 pass
 
-        for sel in ['[class*="price"]', '[data-testid*="price"]',
+        results.sort(key=lambda x: x["preis"])
-                    '.flight-price', 'span[class*="Price"]',
-                    'div[class*="farePrice"]', '[class*="totalPrice"]']:
-            try:
-                elems = sb.find_elements(sel)
-                if elems:
-                    for e in elems[:15]:
-                        p = _parse_preis(e.text)
-                        if p:
-                            results.append({"scanner": "traveloka", "preis": p,
-                                            "waehrung": "EUR", "airline": "",
-                                            "abflug": abflug_iso, "ankunft": rueck_iso,
-                                            "booking_url": booking_url})
-                    if results:
-                        break
-            except Exception:
-                pass
-
-        if not results:
-            for r in _preise_aus_body(body, "traveloka", abflug_iso):
-                r["ankunft"] = rueck_iso
-                r["booking_url"] = booking_url
-                results.append(r)
-
         print(f"[TV] Ergebnis: {[r['preis'] for r in results[:5]]}")
         screenshot_b64 = _take_screenshot(sb)
     return results[:10], screenshot_b64