/hotel_uptime

Primary LanguagePython

Hotel uptime/download calculation

Made using FastAPI and mongodb. Reason to chose mongodb is because it offers fast read speed for large data sets. I chose FastAPI because function in FastAPI are async, meaning it won't block the request thread while the interpolation is running.

How to run

  1. Clone the repo
  2. Create data folder in the root directory with store-status.csv, menu-hours.csv, and bq-results.csv inside it.
  3. Setup environment.
  4. Run python main.py db to populate the database.
  5. Run python main.py to start the server.

Endpoints

http://localhost:8000/api/v1/trigger_report

This endpoint will trigger the report generation. It will return the report id. The report will be generated in the background.

http://localhost:8000/api/v1/get_report/{report_id}

This endpoint will return the report. If the report is not ready, it will send response with status being running. If the report is ready, it will send response with status being complete and the report data.

How interpolation works

Interpolation basically is a way to estimate the value of a function between two known values. In this case, we had to get the store's for every interval. I am using linear interpolation to estimate the value between two known values. The formula is:

y = y1 + (x - x1) * ((y2 - y1) / (x2 - x1))

Where x is the time, y is the store's status, x1 and x2 is the time of the known values, y1 and y2 is the store's status of the known values.

How uptime calculation works

Uptime is calculated by counting the number of up status in the store's status. The uptime is then divided by the total number of store's status. The formula is:

uptime = (up_count / total_count)

Things to improve

  1. Add test
  2. Add more error handling
  3. Cache the report data