Sample GTFS/GTFS Realtime datasets for software testing

Have anyone created a set of sample GTFS or GTFS Realtime datasets used for software unit tests that are publicly shared?

An example of what I'm referring to:

The general idea is to have a sample GTFS dataset that doesn't change, and that models certain conditions (e.g., common errors), so you know that when you make code changes (e.g., in a validator) your expected output is consistent.

Thanks,

Sean

Sean Barbeau

Center for Urban Transportation Research

University of South Florida

Hi Sean,

For anyone interested, I'm tracking these here on the awesome-transit list:

Hi Sean,

 

Yes, we use this type of approach for testing the transitland-lib library that handles GTFS and GTFS Realtime parsing and validation for Transitland:

 

There are test case feeds for all three levels of GTFS static validation:

 

1. "bad entities" at the single row level: https://github.com/interline-io/transitland-lib/tree/master/test/data/bad-entities

 

2. validation errors that that involve entities in one or more files: https://github.com/interline-io/transitland-lib/tree/master/test/data/validator/errors

 

3. best practices: https://github.com/interline-io/transitland-lib/tree/master/test/data/validator/best-practiceshttps://github.com/interline-io/transitland-lib/tree/master/test/data/validator/best-practices

 

You'll notice that it's just the relevant files in each directory for each test case. At the time of running the test, the tooling injects the rest of the feed files that are needed to make the feed complete. In theory, the tooling could be expanded to also generate full feed archives for external use.

 

For GTFS Realtime, we create test cases in JSON and turn them into PB contents in the test rig. Much less coverage for that so far — it's mainly for testing matching logic: https://github.com/interline-io/transitland-server/tree/main/test/data/rt

 Drew

header {
  gtfs_realtime_version: "1.0"
  incrementality: FULL_DATASET
  timestamp: 1668402797
}
entity {
  id: "0"
  vehicle {
    trip {
      trip_id: "54967"
      route_id: "07263"
    }
    position {
      latitude: 25.194374
      longitude: 51.467148
      speed: 0.0
    }
    current_stop_sequence: 15
    current_status: IN_TRANSIT_TO
    congestion_level: STOP_AND_GO
    stop_id: "47027"
    vehicle {
      id: "00750"
      label: "0|0|0|3||0|1|0|1"
      license_plate: "7335"
    }
  }
}
entity {
  id: "1"
  vehicle {
    trip {
      trip_id: "550554"
      route_id: "454545"
    }
    position {
      latitude: 25.15584
      longitude: 51.45545
      speed: 0.0
    }
    current_stop_sequence: 1
    current_status: IN_TRANSIT_TO
    congestion_level: STOP_AND_GO
    stop_id: "48500"
    vehicle {
      id: "00736"
      label: "0|0|0|3||0|1|0|1"
      license_plate: "11233"
    }
  }
}

 https://link.medium.com/iLKViPAkzvb

Then you can use one of the open source tools here to load the GTFS and GTFS real-time data:

Sean