realtime bus trip IDs = garbage?

[Jake Swanson]

Hello again. I'm dipping my toes into the realtime bus endpoint and am surprised at the trip ID values. The realtime bus endpoint has trip IDs ranging from 6852226-6897652. If you examine the current GTFS file, the bus trip IDs currently range from 6945225-6992527. At first glance these IDs are outdated/invalid.

Getting min/max trip IDs for bus routes from current GTFS dataset:

-- route_type=3 is bus routes

> select min(trip_id), max(trip_id) from trips join routes on routes.route_id = trips.route_id where routes.route_type = "3";
6945225|6992527

The realtime bus endpoint I just referenced for this:

http://developer.itsmarta.com/BRDRestService/RestBusRealTimeService/GetAllBus

Is the realtime bus data garbage? Is it using trip IDs from an outdated GTFS schedule?

Your saddened marta.io dev,

Jake Swanson

[Chris Webb]

I am seeing the same thing, and I looked at the old GTFS files on transitfeed, and it appears the trip_ids went from looking like 67xxxxx to 69xxxxx.  I never even saw 68xxxxxx.  I'm attempting to find the best solution for matching the data with this mismatched trip_id.

[Jake Swanson]

I was talking with the MARTA Army folks recently on twitter, essentially asking, "how can do you tie up these garbage trip IDs?" And I think their response boiled down to: Take the general timestamp and try to find the matching trip in the schedule. I'm still trying to imagine this working accurately so maybe I'm missing something. Maybe you can use the BLOCK_ID to help narrow things down, if those don't change across GTFS files?

--
You received this message because you are subscribed to a topic in the Google Groups "atl-transit-developers" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/atl-transit-developers/iGcoQK2Z_q4/unsubscribe.
To unsubscribe from this group and all its topics, send an email to atl-transit-devel...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/atl-transit-developers/4be245e9-33d0-428a-8c3e-ad23bc7cd91f%40googlegroups.com.

[Jake Swanson]

Just wanted to update here: I checked the GTFS today and the TRIP_ID values there are also in the realtime bus output! Here's to hoping they stay that way, if we finish building anything around it.

On Monday, November 18, 2019 at 4:37:38 PM UTC-5, Jake Swanson wrote:

I was talking with the MARTA Army folks recently on twitter, essentially asking, "how can do you tie up these garbage trip IDs?" And I think their response boiled down to: Take the general timestamp and try to find the matching trip in the schedule. I'm still trying to imagine this working accurately so maybe I'm missing something. Maybe you can use the BLOCK_ID to help narrow things down, if those don't change across GTFS files?