COex Backend Test Day

Objectives:

• Create an application in Django using any packages/libraries
• version all source code using GIT
• design a database model to store the imported data in the next step
• Let the categories (Category column) be implemented as a linked model
• create a management command to import data from the supplied file
• create output/listing of internal IDs and names of all locations using
  • a) REST endpoint (preferred - ideally DRF and JSON format)
  • b) Django Views
• add filtering to the page/endpoint by category and whether the location is available with a dog
• create a page/endpoint to display complete data for one location
• create a way to add comments to each location
• create a page with the statistics of the places that are stored in the DB
  • Number of places
  • an overview of the categories and the number of sites in each category
  • the site with the most comments
  • farthest place from the centre of the country
  • number/estimate of English places
• on the page/endpoint listing all places, add the current temperature for each place in degrees Celsius (use API from https://open-meteo.com/ for example)
  • the page should have a mechanism/cache that will query the data only 1 per hour (preventing repeated requests that return the same value)

Assignment Notes:

• admin login via coex:coex
• admin commands are: temperature_update, temperature_update_async, reimport_csv and reimport_csv_async
• you will need your own secret key, just run python utils/new_secret_key.py
• cache refresh is set to 2 hours
• For the purposes of this example only, the admin command temperature_update uses the openmeteo_py library, while temperature_update_async uses the httpx library
• temperature_update_async command is super fast because of async tasks. The sync version took about 6,5 minutes (command temperature_update), basic async version from brach async_temp took more then 2 minutes and the last version from main branch (or from async_test branch) took only 30 seconds!
• reimport_csv_async safe only few second, it depends on hardware and on DB type (sqlite is slow)