Proof of concept serverless app to get the open Citibikes based on lat/long pair. The Citibike GBFS source data is available at https://citibikenyc.com/system-data.
This can be invoked from any client. Siri shortcut is my client of choice, which best supports text/plain.
Send an HTTP GET to http://open-citibikes-lambda.vercel.app/api
Example: run the below command in Terminal.
curl https://open-citibikes-lambda.vercel.app/api
Response is text/plain and will return the number of open Citibike docks using the hardcoded stations in api/index.py.
Example response:
7 Train has 4 open docks
Trader Joes has 10 open docks
48th Ave has 6 open docks
4545 Center has 22 open docks
Court Square S has 4 open docks
Court Square N has 4 open docks
The lambda invocation source code is in api/index.py and will be run every time a GET request is sent to the endpoint. The files in / root: get_hardcode_data.py returns the closest station information given a lat/long pair, which is what is hardcoded into the lambda invocation. The reasoning for this setup is that the GBFS JSON is pretty big, and the retrieval process requires a linear reading of the data as well as a full sorting of all entries, so we want to minimize invocation runtime overhead as much as possible.