This is an example solution for STEP HW6, implemented in Go.

Part 1:

• パタトクカシーー is at http://fantasy-transit-example.appspot.com/pata

Part 2:

• http://fantasy-transit-example.appspot.com/

Ther's two choices for priorities in routing: least transfers, or least stations in the path.

To compute the "least stations" path, it's simply computed via BFS over the train station network.

To compute the "least transfers" path, the train station network graph is transformed into an atlernate graph where all of the stations on the same line share an edge. On this graph, a shortest BFS path finds the stations where transfers should happen. Then between each of these "landmark stations", BFS is used again over the single line used between those "landmark stations" to enumerate a stations between them.

For example this adjacency graph for Tokyo:

And the corresponding line adjacency graph where stations on the same line are connected: