Revolut - Python Developer (Data) Challenge

Programming Tasks

In the beginning we need to create a virtual environment with python 3.6 or higher and install the requirements

$ pip install -r equirements.txt

• Task 1 ^^^^^^^^

All implementation in nest.py file, so we need to run this file.

We have 2 ways to run this file

* from STDIN

    $ cat input.json | python nest.py nesting_level_1 nesting_level_2 ... nesting_level_n

* from file

    $ python nest.py nesting_level_1 nesting_level_2 ... nesting_level_n -f input.json

To get more help, you can use

$ python nest.py -h

• Task 2 ^^^^^^^^

It's a simple API runs over Flask RESTful. All implementation in api.py file. To run the server, please run

$ python api.py

then you can post a request on http://localhost:5000/api/v1/nest with the file in the body of the request Also, you must to add at least one level on the URL, i.e: http://127.0.0.1:5000/api/v1/nest?level=city I implemented a very simple way to authenticate this API, all what you need to do is adding a header Authorization to the request with value Token secret-token-1.

• Unit tests ^^^^^^^^^^^^

I added unit tests for the most of the application, all tests in the folder tests and splits into multiple files. All tests related with python script will be in test_nest.py All tests related with APIs will be in test_apis.py All tests related with SQL Queries will be in test_sql.py

To run unit tests, just run:

$ python -m unittest discover

SQL

All queries you can find it in query.sql file. For the requirements 1 and 2, I added more than a solution to solve it, I tried to do my best to find the most efficient way to solve the problem in a short time. So you can see multiple solutions for each one.

For the thid requirements, I can't find any way to solve it within 5s without editing on the schema, so I edit on it by adding a PK for each table (transactions and exchange_rates) and FK between transactions and exchange_rates, this helped me to solve the problem in ms not seconds! You will find in new_schema.sql, the new tables after editing, and the new scripts for adding multiple data into it. Also, I implemented a trigger function that will run after inserting any transaction to get the er_id for this transaction, you will find this function in new_schema.sql also.