paddington-eurostar/app.py

"""
Combine GWR Bristol→Paddington trains with Eurostar St Pancras→destination trains.
"""

from flask import Flask, render_template, redirect, url_for, request, abort, jsonify
from datetime import date, timedelta
from pathlib import Path
import os

from cache import get_cached, set_cached
import scraper.eurostar as eurostar_scraper
import scraper.gwr_fares as gwr_fares_scraper
import scraper.realtime_trains as rtt_scraper
from trip_planner import combine_trips, find_unreachable_morning_eurostars

RTT_PADDINGTON_URL = (
    "https://www.realtimetrains.co.uk/search/detailed/"
    "gb-nr:PAD/from/gb-nr:{crs}/{date}/0000-2359"
    "?stp=WVS&show=pax-calls&order=wtt"
)

RTT_STATION_URL = (
    "https://www.realtimetrains.co.uk/search/detailed/"
    "gb-nr:{crs}/to/gb-nr:PAD/{date}/0000-2359"
    "?stp=WVS&show=pax-calls&order=wtt"
)

app = Flask(__name__, instance_relative_config=False)
app.config.from_object("config.default")
_local = os.path.join(os.path.dirname(__file__), "config", "local.py")
if os.path.exists(_local):
    app.config.from_pyfile(_local)

import cache
import circle_line

cache.CACHE_DIR = app.config["CACHE_DIR"]
circle_line._TXC_XML = app.config["CIRCLE_LINE_XML"]


def _load_stations():
    tsv = Path(__file__).parent / "data" / "direct_to_paddington.tsv"
    stations = []
    for line in tsv.read_text().splitlines():
        line = line.strip()
        if "\t" in line:
            name, crs = line.split("\t", 1)
            stations.append((name, crs))
    return sorted(stations, key=lambda x: x[0])


STATIONS = _load_stations()
STATION_BY_CRS = {crs: name for name, crs in STATIONS}

DESTINATIONS = {
    "paris": "Paris Gare du Nord",
    "brussels": "Brussels Midi",
    "lille": "Lille Europe",
    "amsterdam": "Amsterdam Centraal",
    "rotterdam": "Rotterdam Centraal",
    "cologne": "Cologne Hbf",
}


@app.route("/")
def index():
    today = date.today().isoformat()
    default_min, default_max = _get_defaults()
    return render_template(
        "index.html",
        destinations=DESTINATIONS,
        today=today,
        stations=STATIONS,
        default_min_connection=default_min,
        default_max_connection=default_max,
        valid_min_connections=sorted(VALID_MIN_CONNECTIONS),
        valid_max_connections=sorted(VALID_MAX_CONNECTIONS),
    )


VALID_MIN_CONNECTIONS = {45, 50, 60, 70, 80, 90, 100, 110, 120}
VALID_MAX_CONNECTIONS = {60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180}


def _get_defaults():
    return (
        app.config["DEFAULT_MIN_CONNECTION"],
        app.config["DEFAULT_MAX_CONNECTION"],
    )


def _parse_connection(raw, default, valid_set):
    try:
        val = int(raw)
    except (TypeError, ValueError):
        return default
    return val if val in valid_set else default


@app.route("/search")
def search():
    slug = request.args.get("destination", "")
    travel_date = request.args.get("travel_date", "")
    station_crs = request.args.get("station_crs", "BRI")
    if station_crs not in STATION_BY_CRS:
        station_crs = "BRI"
    default_min, default_max = _get_defaults()
    min_conn = _parse_connection(
        request.args.get("min_connection"), default_min, VALID_MIN_CONNECTIONS
    )
    max_conn = _parse_connection(
        request.args.get("max_connection"), default_max, VALID_MAX_CONNECTIONS
    )
    if slug in DESTINATIONS and travel_date:
        return redirect(
            url_for(
                "results",
                station_crs=station_crs,
                slug=slug,
                travel_date=travel_date,
                min_connection=None if min_conn == default_min else min_conn,
                max_connection=None if max_conn == default_max else max_conn,
            )
        )
    return redirect(url_for("index"))


@app.route("/results/<station_crs>/<slug>/<travel_date>")
def results(station_crs, slug, travel_date):
    departure_station_name = STATION_BY_CRS.get(station_crs)
    if departure_station_name is None:
        abort(404)
    destination = DESTINATIONS.get(slug)
    if not destination or not travel_date:
        return redirect(url_for("index"))

    default_min, default_max = _get_defaults()
    min_connection = _parse_connection(
        request.args.get("min_connection"), default_min, VALID_MIN_CONNECTIONS
    )
    max_connection = _parse_connection(
        request.args.get("max_connection"), default_max, VALID_MAX_CONNECTIONS
    )

    # Redirect to clean URL when both params are at their defaults
    if (
        "min_connection" in request.args or "max_connection" in request.args
    ) and min_connection == default_min and max_connection == default_max:
        return redirect(
            url_for("results", station_crs=station_crs, slug=slug, travel_date=travel_date)
        )

    user_agent = request.headers.get("User-Agent", rtt_scraper.DEFAULT_UA)

    rtt_cache_key = f"rtt_{station_crs}_{travel_date}"
    es_cache_key = f"eurostar_{travel_date}_{destination}"
    gwr_fares_cache_key = f"gwr_fares_{station_crs}_{travel_date}"
    gwr_advance_cache_key = f"gwr_advance_{station_crs}_{travel_date}"

    cached_rtt = get_cached(rtt_cache_key)
    cached_es = get_cached(es_cache_key, ttl=24 * 3600)
    cached_gwr_fares = get_cached(gwr_fares_cache_key, ttl=30 * 24 * 3600)
    cached_advance_fares = get_cached(gwr_advance_cache_key, ttl=24 * 3600)
    from_cache = bool(cached_rtt and cached_es)

    error = None

    if cached_rtt:
        gwr_trains = cached_rtt
    else:
        try:
            gwr_trains = rtt_scraper.fetch(travel_date, user_agent, station_crs)
            set_cached(rtt_cache_key, gwr_trains)
        except Exception as e:
            gwr_trains = []
            error = f"Could not fetch GWR trains: {e}"

    if cached_es:
        eurostar_services = cached_es
    else:
        try:
            eurostar_services = eurostar_scraper.fetch(destination, travel_date)
            set_cached(es_cache_key, eurostar_services)
        except Exception as e:
            eurostar_services = []
            msg = f"Could not fetch Eurostar times: {e}"
            error = f"{error}; {msg}" if error else msg

    if cached_gwr_fares:
        gwr_fares = cached_gwr_fares
    else:
        try:
            gwr_fares = gwr_fares_scraper.fetch(station_crs, travel_date)
            set_cached(gwr_fares_cache_key, gwr_fares)
        except Exception as e:
            gwr_fares = {}
            msg = f"Could not fetch GWR fares: {e}"
            error = f"{error}; {msg}" if error else msg

    eurostar_trains = eurostar_services
    eurostar_prices = {
        s["depart_st_pancras"]: {
            "price":       s.get("price"),
            "seats":       s.get("seats"),
            "plus_price":  s.get("plus_price"),
            "plus_seats":  s.get("plus_seats"),
        }
        for s in eurostar_services
    }

    trips = combine_trips(
        gwr_trains,
        eurostar_trains,
        travel_date,
        min_connection,
        max_connection,
        gwr_fares,
    )

    # Annotate each trip with Eurostar prices and total cost (walk-on + standard)
    for trip in trips:
        es = eurostar_prices.get(trip["depart_st_pancras"], {})
        es_price = es.get("price")
        trip["eurostar_price"] = es_price
        trip["eurostar_seats"] = es.get("seats")
        trip["eurostar_plus_price"] = es.get("plus_price")
        trip["eurostar_plus_seats"] = es.get("plus_seats")
        gwr_p = trip.get("ticket_price")
        circle_svcs = trip.get("circle_services")
        circle_fare = circle_svcs[0]["fare"] if circle_svcs else 0
        trip["total_price"] = (
            gwr_p + es_price + circle_fare
            if (gwr_p is not None and es_price is not None)
            else None
        )

    # If the API returned journeys but every price is None, tickets aren't on sale yet
    no_prices_note = None
    if eurostar_prices and all(
        v.get("price") is None for v in eurostar_prices.values()
    ):
        no_prices_note = (
            "Eurostar prices not yet available — tickets may not be on sale yet."
        )

    unreachable_morning_services = find_unreachable_morning_eurostars(
        gwr_trains,
        eurostar_trains,
        travel_date,
        min_connection,
        max_connection,
    )
    for svc in unreachable_morning_services:
        es = eurostar_prices.get(svc["depart_st_pancras"], {})
        svc["eurostar_price"] = es.get("price")
        svc["eurostar_seats"] = es.get("seats")
        svc["eurostar_plus_price"] = es.get("plus_price")
        svc["eurostar_plus_seats"] = es.get("plus_seats")

    # Only keep unreachable services that depart before the first reachable Eurostar.
    # Services after the first reachable one are omitted (they aren't "Too early").
    if trips:
        first_es_depart = min(t["depart_st_pancras"] for t in trips)
        unreachable_morning_services = [
            s
            for s in unreachable_morning_services
            if s["depart_st_pancras"] < first_es_depart
        ]

    result_rows = sorted(
        [{"row_type": "trip", **trip} for trip in trips]
        + [
            {"row_type": "unreachable", **service}
            for service in unreachable_morning_services
        ],
        key=lambda row: row["depart_st_pancras"],
    )

    dt = date.fromisoformat(travel_date)
    prev_date = (dt - timedelta(days=1)).isoformat()
    next_date = (dt + timedelta(days=1)).isoformat()
    travel_date_display = dt.strftime("%A %-d %B %Y")

    eurostar_url = eurostar_scraper.search_url(destination, travel_date)
    rtt_url = RTT_PADDINGTON_URL.format(crs=station_crs, date=travel_date)
    rtt_station_url = RTT_STATION_URL.format(crs=station_crs, date=travel_date)

    url_min = None if min_connection == default_min else min_connection
    url_max = None if max_connection == default_max else max_connection

    return render_template(
        "results.html",
        trips=trips,
        result_rows=result_rows,
        unreachable_morning_services=unreachable_morning_services,
        destinations=DESTINATIONS,
        destination=destination,
        travel_date=travel_date,
        slug=slug,
        station_crs=station_crs,
        departure_station_name=departure_station_name,
        prev_date=prev_date,
        next_date=next_date,
        travel_date_display=travel_date_display,
        gwr_count=len(gwr_trains),
        eurostar_count=len(eurostar_trains),
        from_cache=from_cache,
        error=error,
        no_prices_note=no_prices_note,
        eurostar_url=eurostar_url,
        rtt_url=rtt_url,
        rtt_station_url=rtt_station_url,
        min_connection=min_connection,
        max_connection=max_connection,
        default_min_connection=default_min,
        default_max_connection=default_max,
        url_min_connection=url_min,
        url_max_connection=url_max,
        cached_advance_fares=cached_advance_fares,
        valid_min_connections=sorted(VALID_MIN_CONNECTIONS),
        valid_max_connections=sorted(VALID_MAX_CONNECTIONS),
    )


@app.route("/api/advance_fares/<station_crs>/<travel_date>")
def api_advance_fares(station_crs, travel_date):
    if station_crs not in STATION_BY_CRS:
        abort(404)
    cache_key = f"gwr_advance_{station_crs}_{travel_date}"
    cached = get_cached(cache_key, ttl=24 * 3600)
    if cached is not None:
        return jsonify(cached)
    try:
        fares = gwr_fares_scraper.fetch_advance(station_crs, travel_date)
        set_cached(cache_key, fares)
        return jsonify(fares)
    except Exception as e:
        return jsonify({"error": str(e)}), 500


if __name__ == "__main__":
    app.run(debug=True, host="0.0.0.0")