Fix Guestline + Mews direct scrapers after engine API changes

Both competitor booking engines changed their API contracts, silently
breaking their direct scrapes ~16 days ago (200 responses with empty
bodies, so no exception was ever raised):

- Guestline: the /enhanced availability endpoint now returns a bare []
  for every date. Switch to the base /api/availabilities/{coll}/{hotel}
  endpoint, which returns {"rooms":[...]} with the same room shape.
- Mews: getAvailability now rejects the old body with "Invalid
  EnterpriseId" — it requires enterpriseId + serviceId. getPricing
  additionally requires currencyCode. Capture the enterprise's
  defaultCurrencyCode in _ensure_config and send all three.

Also stop last_scraped_at advancing when a run saves 0 rows, so the
overview no longer reports a phantom-fresh scrape while the per-date
data stays stale — the symptom that surfaced this.

Verified live: Three Ways now yields 41 rows/night, Old Stocks 10.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
This commit is contained in:
jtricerolph 2026-08-13 14:09:13 +00:00
parent 22e3670531
commit b916f1a09b
3 changed files with 31 additions and 14 deletions

View file

@ -39,11 +39,11 @@ class GuestlineProfile(BaseProfile):
hotel_id = params["hotel_id"] hotel_id = params["hotel_id"]
collection_id = params.get("collection_id", "MT") collection_id = params.get("collection_id", "MT")
dep = (date.fromisoformat(arrival) + timedelta(days=nights)).isoformat() dep = (date.fromisoformat(arrival) + timedelta(days=nights)).isoformat()
url = f"https://booking.eu.guestline.app/api/availabilities/{collection_id}/{hotel_id}/enhanced" url = f"https://booking.eu.guestline.app/api/availabilities/{collection_id}/{hotel_id}"
r = await client.get(url, params={ r = await client.get(url, params={
"arrival": arrival, "departure": dep, "adults": 2, "children": 0, "arrival": arrival, "departure": dep, "adults": 2, "children": 0,
}, headers=HEADERS, timeout=20) }, headers=HEADERS, timeout=20)
if r.status_code == 404: if r.status_code == 404:
return [] return []
r.raise_for_status() r.raise_for_status()
return r.json().get("availabilities", {}).get("rooms", []) return r.json().get("rooms", [])

View file

@ -58,7 +58,11 @@ async def _ensure_config(client, params: dict) -> bool:
if not enterprise_id: if not enterprise_id:
enterprise_id = (data.get("enterprises") or [{}])[0].get("id", "") enterprise_id = (data.get("enterprises") or [{}])[0].get("id", "")
tz_name = (data.get("enterprises") or [{}])[0].get("ianaTimeZoneIdentifier", "Europe/London") enterprises = data.get("enterprises") or []
ent_obj = next((e for e in enterprises if e.get("id") == enterprise_id),
enterprises[0] if enterprises else {})
tz_name = ent_obj.get("ianaTimeZoneIdentifier", "Europe/London")
currency = ent_obj.get("defaultCurrencyCode") or "GBP"
# Adult age category: classification=="Adult" or no maximalAge, filtered to this service # Adult age category: classification=="Adult" or no maximalAge, filtered to this service
age_cat_id = "" age_cat_id = ""
@ -72,6 +76,7 @@ async def _ensure_config(client, params: dict) -> bool:
params["enterprise_id"] = enterprise_id params["enterprise_id"] = enterprise_id
params["service_id"] = service_id params["service_id"] = service_id
params["age_category_id"] = age_cat_id params["age_category_id"] = age_cat_id
params["currency"] = currency
params["_tz"] = ZoneInfo(tz_name) params["_tz"] = ZoneInfo(tz_name)
return bool(enterprise_id and service_id) return bool(enterprise_id and service_id)
@ -134,6 +139,8 @@ class MewsProfile(BaseProfile):
f"{API_BASE}/services/getAvailability", f"{API_BASE}/services/getAvailability",
json={ json={
"bookingEngineId": params["booking_engine_id"], "bookingEngineId": params["booking_engine_id"],
"enterpriseId": params["enterprise_id"],
"serviceId": params["service_id"],
"startUtc": _midnight_utc(today, tz), "startUtc": _midnight_utc(today, tz),
"endUtc": _midnight_utc(today + timedelta(days=DAYS_AHEAD), tz), "endUtc": _midnight_utc(today + timedelta(days=DAYS_AHEAD), tz),
"client": CLIENT, "client": CLIENT,
@ -169,6 +176,9 @@ class MewsProfile(BaseProfile):
f"{API_BASE}/services/getPricing", f"{API_BASE}/services/getPricing",
json={ json={
"configurationId": params["booking_engine_id"], "configurationId": params["booking_engine_id"],
"enterpriseId": params["enterprise_id"],
"serviceId": params["service_id"],
"currencyCode": params.get("currency", "GBP"),
"startUtc": _midnight_utc(arrival_date, tz), "startUtc": _midnight_utc(arrival_date, tz),
"endUtc": _midnight_utc(arrival_date + timedelta(days=nights), tz), "endUtc": _midnight_utc(arrival_date + timedelta(days=nights), tz),
"occupancyData": [{"ageCategoryId": age_cat_id, "personCount": 2}], "occupancyData": [{"ageCategoryId": age_cat_id, "personCount": 2}],

View file

@ -123,19 +123,25 @@ def run_scrape(hotel_id: int, profile_name: str, params: dict):
loop = asyncio.new_event_loop() loop = asyncio.new_event_loop()
try: try:
loop.run_until_complete(_run_scrape_async(hotel_id, profile, params, scraped_at)) rows_saved = loop.run_until_complete(_run_scrape_async(hotel_id, profile, params, scraped_at))
finally: finally:
loop.close() loop.close()
db = SyncSessionLocal() # Only advance last_scraped_at when the run actually captured data. A run that
try: # saved nothing means the engine returned empty/errored — don't let the overview
db.execute( # report a phantom-fresh scrape while the per-date data stays stale.
text("UPDATE direct_competitor_hotels SET last_scraped_at = :ts WHERE id = :id"), if rows_saved:
{"ts": scraped_at, "id": hotel_id} db = SyncSessionLocal()
) try:
db.commit() db.execute(
finally: text("UPDATE direct_competitor_hotels SET last_scraped_at = :ts WHERE id = :id"),
db.close() {"ts": scraped_at, "id": hotel_id}
)
db.commit()
finally:
db.close()
else:
log.warning(f"Hotel {hotel_id}: scrape saved 0 rows — last_scraped_at left unchanged")
async def _run_scrape_async(hotel_id: int, profile, params: dict, scraped_at: datetime): async def _run_scrape_async(hotel_id: int, profile, params: dict, scraped_at: datetime):
@ -144,7 +150,7 @@ async def _run_scrape_async(hotel_id: int, profile, params: dict, scraped_at: da
arrival_dates = await profile.fetch_arrival_dates(client, params) arrival_dates = await profile.fetch_arrival_dates(client, params)
except Exception as e: except Exception as e:
log.error(f"Hotel {hotel_id}: failed to fetch arrival dates: {e}") log.error(f"Hotel {hotel_id}: failed to fetch arrival dates: {e}")
return return 0
log.info(f"Hotel {hotel_id}: {len(arrival_dates)} bookable dates") log.info(f"Hotel {hotel_id}: {len(arrival_dates)} bookable dates")
@ -305,3 +311,4 @@ async def _run_scrape_async(hotel_id: int, profile, params: dict, scraped_at: da
db.close() db.close()
log.info(f"Hotel {hotel_id}: scrape complete, {rows_saved} rows saved") log.info(f"Hotel {hotel_id}: scrape complete, {rows_saved} rows saved")
return rows_saved