train-api

Work in progress. "train-api" is a placeholder name; the final name isn't chosen yet.

Open train data for rail passes: MAX JEUNE, MAX SENIOR and Interrail, one simple API per operator, built from each operator's own open timetable:

Operator	Source	Licence	Passes
sncf	SNCF Open Data tgvmax (TGV INOUI, Intercités)	Licence Ouverte	MAX JEUNE, MAX SENIOR, Interrail
renfe	Renfe Data, high-speed, long and medium distance (GTFS)	CC BY 4.0	Interrail
nl	Dutch national timetable via OVapi (GTFS): international and Intercity trains	CC0	Interrail
de	DELFI via gtfs.de (GTFS): German long-distance trains (ICE, IC, EC…)	CC BY 4.0	Interrail
it	Trenitalia via Italy's National Access Point, converted to GTFS by deryclem/trenitalia-gtfs	CC BY 4.0	Interrail
ch	opentransportdata.swiss (GTFS): Swiss long-distance trains, IR and RE between their stations	free use, source cited	Interrail
no	Entur (GTFS): Vy, Go-Ahead and SJ trains in Norway	NLOD	Interrail
fi	Fintraffic / digitraffic.fi (GTFS): VR long-distance trains	CC BY 4.0	Interrail
ie	National Transport Authority (GTFS): Irish Rail, without the DART	CC BY 4.0	Interrail
se	Samtrafiken via Trafiklab, GTFS Sverige 2: long-distance trains, regional between their stations. Needs TRAFIKLAB_KEY	CC0	Interrail
gb	Network Rail open data SCHEDULE feed: express trains and sleepers. Needs NETWORK_RAIL_EMAIL and NETWORK_RAIL_PASSWORD	Network Rail licence (attribution)	Interrail

se and gb come from feeds behind a free sign-up: the publish workflow reads the keys from repository secrets and leaves an operator out while its secret isn't set.

The SNCF part is built on top of MAX-Finder: the same feed, without MAX-Finder's "free MAX seat only" filter, so every train an Interrail holder can book shows up. Each operator is published on its own; trips that change between operators aren't joined here.

No open data says whether an Interrail pass-holder seat is left, so Interrail trains are published as running, with the seat unknown.

Use the API

Free, no key, CORS open, refreshed daily: start with the developer guide (endpoints, JS and Python examples, licences and credits, stability promise) and the OpenAPI spec.

How it works

1. npm run fetch downloads every source to data/raw/: the SNCF tgvmax feed (every TGV INOUI and Intercités train for ~30 days) and each GTFS operator's zip, unzipped.
2. npm run build:api maps it to the open train format and writes the static API to public/v1/ in the compact snapshot format: each service is stored once with the dates it runs, about 20x smaller than the raw rows.
3. The Publish API workflow does both daily and deploys public/ to GitHub Pages. There is no server and no cost.

Live at https://offware-apps.github.io/train-api/v1/index.json once Pages is on (Settings → Pages → Source: GitHub Actions).

Develop

npm ci
npm run check                                              # typecheck + tests
npm run build:api -- tests/fixtures/sncf-tgvmax.sample.json \
  --gtfs renfe=tests/fixtures/gtfs-mini --gb tests/fixtures/gb-schedule.json \
  --from 2026-10-05                                          # build from the fixtures

Next

• The front-end on top of MAX-Finder's search core, with Interrail as the pass.
• More operators through the same GTFS reader. Waiting on a free sign-up or a licence check: Belgium (SNCB), Austria (ÖBB), Denmark (Rejseplanen licence).

Data: SNCF Open Data (Licence Ouverte), Renfe Data (CC BY 4.0), OVapi / NDOV (CC0), DELFI e.V. / gtfs.de (CC BY 4.0), Trenitalia via the Italian NAP and Clément Desouche (CC BY 4.0), opentransportdata.swiss, Entur (NLOD), Fintraffic / digitraffic.fi (CC BY 4.0), National Transport Authority (CC BY 4.0), Samtrafiken (CC0), contains information of Network Rail Infrastructure Limited. Code: AGPL-3.0.