In this repo I show how to log in and get list of rentals from nextbike API. I ride it a lot and wanted to have some statistics.
I took inspiration from other nextbike projects:
- https://github.com/bdmbdsm/nextbike_api
- https://github.com/cybre-finn/nextbike-api-reverse-engineering
login.py will print login response from which you can find a "login key", and then use it with get_list.py.
Run as MOBILE="+420735589654" PIN=232523 ./login.py and find login_key in the output.
get_list.py will download list of your activities to list.json. You can then analyze the file.
Run as LOGIN_KEY=sg9032rj32r09rj3 ./get_list.py and see file list.json.
The downloaded json document contains map with keys server_time, user and account. account is a map containing key items which is a list of "nodes", maps of types voucher, payment and rental. rental nodes are data about pickup and return of a bike. Those are the interesting part.
The rental nodes are maps with keys like start_time, start_place_latitude, start_place_longitude, end_time and so on. There's mode, but the time and return and end places of rentals are relevant for my analysis.
The analysis filters rentals from the list.json and shows totals, month and day aggregation, and a map. I first process list.json into a dataframe which I saved as df.h5 in the repo, so that I don't have to put my list.json here.
- Total distance travelled: 1041 km [0]
- Total time spent: 77 hours
- Total trips: 409
[0] I measured that my most common ride is 3.01 km aerial and 3.8 km on the map, so I factor all measured distances by 1.25