reStations

This is an implementation of the OSDM API spec's Places endpoint, backed by Trainline EU's stations dataset.

The easiest way to use reStations is as a Docker container using the image that we published including the dataset:

docker run -p 3000:3000 --rm mainmatter/restations
curl localhost:3000/places

New versions of the image are published regularly as the dataset is updated.

Endpoints supported

GET /places

Request all places:

curl localhost:3000/places

GET /places/{id}

Example: fetch Lisboa Santa Apolónia station with its UIC:

curl localhost:3000/places/8721428

POST /places (search)

By name

Example: search stations with Lisboa in its name.

Works with both Portuguese and English versions, and in other languages as well. See Trainline's data repository for the complete list of supported languages.

curl -X POST -H "Content-Type: application/json" \
-d '{"placeInput": {"name": "Lisboa"}}' \
localhost:3000/places

By geo position

Example: search stations near Lisboa Santa Apolónia station:

curl -X POST -H "Content-Type: application/json" \
-d '{"placeInput": {"geoPosition": {"latitude": 38.71387, "longitude": -9.122271}}}' \
localhost:3000/places

Restrictions

We currently support requesting the number of results to be returned.

curl -X POST -H "Content-Type: application/json" \
-d '{"placeInput": {"name": "London"}, "restrictions": {"numberOfResults": 3}}' \
localhost:3000/places

This will return St Pancras, Liverpool Street and Blackfriars stations:

{
  "places": [
    {
      "id": "urn:uic:stn:7015400",
      "objectType": "StopPlace",
      "name": "London St Pancras International",
      "alternativeIds": [],
      "geoPosition": {
        "latitude": 51.531921,
        "longitude": -0.126361
      },
      "countryCode": "GB",
      "links": []
    },
    {
      "id": "urn:uic:stn:7069650",
      "objectType": "StopPlace",
      "name": "London Liverpool Street",
      "alternativeIds": [],
      "geoPosition": {
        "latitude": 51.517551,
        "longitude": -0.08021
      },
      "countryCode": "GB",
      "links": []
    },
    {
      "id": "urn:uic:stn:7051120",
      "objectType": "StopPlace",
      "name": "London Blackfriars",
      "alternativeIds": [],
      "geoPosition": {
        "latitude": 51.510735,
        "longitude": -0.103554
      },
      "countryCode": "GB",
      "links": []
    }
  ]
}

Support for other restrictions request fields for POST /places is in the works.

Working with reStations

reStations can also be used directly as a Rust project. To run the project, prepare the database first:

cargo db create

This is important since reStations uses sqlx which does compile-time checks on the database schema and needs the DATABASE_URL environment variable to be able to connect to the database.

To synchronize the data into the local database from the Trainline EU station dataset, run

cargo db sync

Then run the applications from the project root:

cargo run

Prerequisites

• Rust (install via rustup)
• sqlite3 (see SQLite)

Project Structure

Distinct parts of the project are separated into separate crates:

.
├── cli    // CLI tools for generating project files
├── config // Defines the `Config` struct and handles building the configuration from environment-specific TOML files and environment variables
├── macros // Contains macros for application tests
└── web    // The web interface as well as tests for it

Environment

The project uses .env and .env.test files to store configuration settings for the development and test environments respectively. Those files are read automatically by the parts of the application that require configuration settings to be present.

Commands

Running the application in development mode:

cargo run

Running the application tests:

cargo db create -e test
cargo test

Generating project files like entities, controllers, tests, etc. (see the CLI create for detailed documentation):

cargo generate

Building the project's docs:

Building documentation

Build the project's documentation with:

cargo doc --workspace --all-features

License

Copyright © 2025- Mainmatter GmbH (https://mainmatter.com), released under the MIT and Apache licenses.

Trainline EU's stations dataset is released under the Open Data Commons Open Database License v1.0 license which allows private and commercial use.