parse subway gtfs

subway.py

import requests

key = '' 
r = requests.get('http://datamine.mta.info/mta_esi.php?key=%s&feed_id=1'%(key))

# http://datamine.mta.info/sites/all/files/pdfs/nyct-subway.proto.txt
# https://developers.google.com/transit/gtfs-realtime/gtfs-realtime-proto
# https://github.com/google/protobuf

import gtfs_realtime_pb2, nyct_subway_pb2

# http://datamine.mta.info/sites/all/files/pdfs/GTFS-Realtime-NYC-Subway%20version%201%20dated%207%20Sep.pdf
# https://groups.google.com/forum/#!topic/mtadeveloperresources/PeWNe4WG890
subways = gtfs_realtime_pb2.FeedMessage()

subways.ParseFromString(r.content)
subways.entity

for T in subways.entity:
    print T

trips.py

# stick into cron, hopefully doesn't crash!

[(
    str(E.trip_update.trip.route_id), 
    len(E.trip_update.stop_time_update),
    datetime.datetime.fromtimestamp(
        int(E.trip_update.stop_time_update[0].arrival.time)
    ).strftime('%H:%M:%S')
    , 
    str(E.trip_update.stop_time_update[0].stop_id),
    str(E.trip_update.trip.trip_id)
) for E in subways.entity if len(E.trip_update.stop_time_update) > 0]

"""
[('1', 1, '10:00:33', '140S', '047600_1..S'),
 ('1', 25, '10:02:19', '116S', '051600_1..S'),
 ('1', 2, '09:58:31', '139S', '048150_1..S07R'),
 ('1', 4, '10:01:45', '137S', '048400_1..S'),
 ('1', 6, '10:00:28', '135S', '048600_1..S02R'),
 ('1', 5, '10:01:54', '107N', '048950_1..N02R'),
 ('1', 8, '10:00:48', '133S', '049100_1..S02R'),
 ('1', 9, '10:02:21', '111N', '049400_1..N02R'),
..."""

update.gtfs

id: "000002"
vehicle {
  trip {
    trip_id: "078200_1..N02R"
    start_date: "20141217"
    route_id: "1"
    [nyct_trip_descriptor] {
      train_id: "01 1302  SFY/242"
      is_assigned: true
      direction: NORTH
    }
  }
  current_stop_sequence: 38
  current_status: IN_TRANSIT_TO
  timestamp: 1418842977
  stop_id: "101N"
}