Add airbnb trip detail parser.

parse_airbnb.py

@@ -0,0 +1,210 @@
+#!/usr/bin/python3
+
+import json
+import re
+import sys
+import typing
+from datetime import datetime
+from typing import Any
+from zoneinfo import ZoneInfo
+
+import lxml.html
+import pycountry
+import yaml
+
+StrDict = dict[str, typing.Any]
+
+
+def build_datetime(date_str: str, time_str: str, tz_name: str) -> datetime:
+    """
+    Combine an ISO date string, HH:MM time string, and a timezone name
+    into a timezone-aware datetime in the specified timezone.
+    """
+    dt_str = f"{date_str}T{time_str}"
+    naive_dt = datetime.fromisoformat(dt_str)
+    return naive_dt.replace(tzinfo=ZoneInfo(tz_name))
+
+
+def list_to_dict(items: list) -> dict[str, int]:
+    """Convert a flat list to a dict, assuming alternating keys and values."""
+    return {items[i]: items[i + 1] for i in range(0, len(items), 2)}
+
+
+def extract_country_code(address: str) -> str | None:
+    """Return ISO 3166-1 alpha-2 country code from a free-text address."""
+    address_lower = address.lower()
+    for country in pycountry.countries:
+        if country.name.lower() in address_lower:
+            return str(country.alpha_2.lower())
+        if (
+            hasattr(country, "official_name")
+            and country.official_name.lower() in address_lower
+        ):
+            return str(country.alpha_2.lower())
+    return None
+
+
+def get_json_blob(tree) -> str:
+    data_id = "data-injector-instances"
+    js_string = tree.xpath(f'//*[@id="{data_id}"]/text()')[0]
+    return js_string
+
+
+def get_ui_state(tree) -> StrDict:
+    data_id = "data-injector-instances"
+    js_string = tree.xpath(f'//*[@id="{data_id}"]/text()')[0]
+    # print(js_string)
+    # sys.exit(0)
+    big_blob = json.loads(js_string)
+    ui_state = walk_tree(big_blob, "uiState")
+    # print(json.dumps(ui_state))
+
+    return list_to_dict(ui_state[0])
+
+
+def get_reservation_data(ui_state: StrDict) -> StrDict:
+    return {
+        row["id"]: row for row in ui_state["reservation"]["scheduled_event"]["rows"]
+    }
+
+
+def get_room_url(tree) -> str:
+    for e in tree.xpath('//a[@data-testid="reservation-destination-link"]'):
+        href = e.get("href")
+        assert isinstance(href, str)
+        if not href.startswith("/room"):
+            continue
+        return "https://www.airbnb.co.uk" + href
+
+
+def get_price_from_reservation(reservation: StrDict) -> str:
+    price_string = reservation["payment_summary"]["subtitle"]
+    assert isinstance(price_string, str)
+    tc = "Total cost: "
+    assert price_string.startswith(tc)
+    price = price_string[len(tc) :]
+    assert price[0] == "£"
+    return price[1:]
+
+
+def extract_booking_from_html(html_file: str) -> StrDict:
+    """Extract booking information from Airbnb HTML file."""
+
+    with open(html_file, "r", encoding="utf-8") as f:
+        text_content = f.read()
+
+    confirmation_code = re.search(
+        r"/trips/v1/reservation-details/ro/RESERVATION2_CHECKIN/([A-Z0-9]+)",
+        text_content,
+    ).group(1)
+
+    tree = lxml.html.parse(html_file)
+    root = tree.getroot()
+    try:
+        ui_state = get_ui_state(tree)
+    except Exception:
+        print(html_file)
+        raise
+    # print(json.dumps(ui_state))
+
+    reservation = get_reservation_data(ui_state)
+    m_guests = re.match(r"^(\d+) guests?$", reservation["guests"]["subtitle"])
+    number_of_adults = int(m_guests.group(1))
+
+    price = get_price_from_reservation(reservation)
+    # print(reservation["payment_summary"]["subtitle"])
+    # print(json.dumps(reservation))
+    metadata = ui_state["reservation"]["metadata"]
+    country_code = metadata["country"].lower()
+    # pprint(metadata)
+
+    # print(json.dumps(x))
+
+    title = reservation["dynamic_marquee_title_image_v3"]["title"]
+    location = title.rpartition(" in ")[2]
+
+    # print(json.dumps(reservation))
+    checkin_checkout = reservation["checkin_checkout_arrival_guide"]
+    # pprint(checkin_checkout)
+    check_in_time = checkin_checkout["leading_subtitle"]
+    check_out_time = checkin_checkout["trailing_subtitle"]
+
+    check_in = build_datetime(
+        metadata["check_in_date"], check_in_time, metadata["timezone"]
+    )
+
+    check_out = build_datetime(
+        metadata["check_out_date"], check_out_time, metadata["timezone"]
+    )
+
+    # print(check_in, check_out)
+
+    address = reservation["map"]["address"]
+
+    # country_code = extract_country_code(address)
+
+    # if "header_action.pdp" not in reservation:
+    #     pprint(reservation)
+
+    if "header_action.pdp" in reservation:
+        name = reservation["header_action.pdp"]["subtitle"]
+    else:
+        name = root.findtext(".//h1")
+
+    booking = {
+        "type": "apartment",
+        "operator": "airbnb",
+        "name": name,
+        "location": location,
+        "booking_reference": confirmation_code,
+        "booking_url": f"https://www.airbnb.co.uk/trips/v1/reservation-details/ro/RESERVATION2_CHECKIN/{confirmation_code}",
+        "address": address,
+        "country": country_code,
+        "latitude": metadata["lat"],
+        "longitude": metadata["lng"],
+        "timezone": metadata["timezone"],
+        "from": check_in,
+        "to": check_out,
+        "price": price,
+        "currency": "GBP",
+        "number_of_adults": number_of_adults,
+    }
+
+    booking["url"] = get_room_url(tree)
+
+    return booking
+
+
+def walk_tree(data: Any, want_key: str) -> Any:
+    """Recursively search for a dict containing 'reservation' and return its value."""
+    if isinstance(data, dict):
+        if want_key in data:
+            return data[want_key]
+        for key, value in data.items():
+            result = walk_tree(value, want_key)
+            if result is not None:
+                return result
+    elif isinstance(data, list):
+        for item in data:
+            result = walk_tree(item, want_key)
+            if result is not None:
+                return result
+    return None
+
+
+def main() -> None:
+    """Main function."""
+
+    filenames = sorted(sys.argv[1:])
+
+    bookings = []
+    for html_file in filenames:
+        booking = extract_booking_from_html(html_file)
+        assert booking
+        bookings.append(booking)
+
+    print(yaml.dump(bookings, sort_keys=False))
+
+
+if __name__ == "__main__":
+    main()