This is the preprocessor for an exploration of commutes & rent in London.

It has three purposes:

• It sucks up Transport For London timetable data and uses it to generate a large number of JSON files which encode what time you'd have to leave each station in the Underground network in order to arrive at each other station by a specified time.

This is done by processing the timetable data into a transit graph, then carrying out a dynamic programming algorithm strongly reminiscent of Dijkstra's (supplemented with a few culling heuristics) for each destination station.

• It uses the station location information from TFL along with postcode and rent information from the Office of National Statistics to produce a JSON file that maps stations to the rent statistics of the Local Authority District they lie in.

• It uses the station location information (along with a hand-crafted file of corrections) to produce a JSON file that maps station names from the TfL data to coordinates.

Using these three sets of files, the frontend can generate a map of the stations and a chart of travel times between each and the rents in the areas around them.