Data transformations with Python

This coding challenge is a collection of Python jobs that are supposed to extract, transform and load data. These jobs are using PySpark to process larger volumes of data and are supposed to run on a Spark cluster (via spark-submit).

Gearing Up for the Pairing Session

✅ Goals

1. Get a working environment See local local
2. Get a high-level understanding of the code and test dataset structure
3. Have your preferred text editor or IDE setup and ready to go.

❌ Non-Goals

• solving the exercises / writing code

  ⚠️ The exercises will be given at the time of interview, and solved by pairing with the interviewer.

Local Setup

Pre-requisites

Please make sure you have the following installed and can run them

• Python (3.13.X), you can use for example pyenv to manage your python versions locally
• Poetry
• Java (17), you can use sdkman to install and manage java locally

Windows users

We recommend using WSL 2 on Windows for this exercise, due to the lack of support of windows paths from Hadoop/Spark.

Follow instructions on the Windows official page and then the linux install.

Install all dependencies

poetry install

Verify setup

All of the following commands should be running successfully

Run unit tests

poetry run pytest tests/unit

Run integration tests

poetry run pytest tests/integration

Run style checks

poetry run mypy --ignore-missing-imports --disallow-untyped-calls --disallow-untyped-defs --disallow-incomplete-defs \
            data_transformations tests

poetry run ruff format && poetry run ruff check

Anything else?

All commands are passing?
You are good to go!

⚠️ do not try to solve the exercises ahead of the interview

You are allowed to customize your environment (having the test in vscode directly for example): feel free to spend the time making this comfortable for you. This is not an expectation.

Jobs

There are two exercises in this repo: Word Count, and Citibike.

Currently, these exist as skeletons, and have some initial test cases which are defined but some are skipped.

The following section provides context over them.

⚠️ do not try to solve the exercises ahead of the interview

Code walk

/
├─ /data_transformations # Contains the main python library
│ # with the code to the transformations
│
├─ /jobs # Contains the entry points to the jobs
│ # performs argument parsing, and are
│ # passed to `spark-submit`
│
├─ /resources # Contains the raw datasets for the jobs
│
├─ /tests
│ ├─ /units # contains basic unit tests for the code
│ └─ /integration # contains integrations tests for the jobs
│ # and the setup
│
├─ .gitignore
├─ .pylintrc # configuration for pylint
├─ LICENCE
├─ poetry.lock
├─ pyproject.toml
└─ README.md # The current file

Word Count

A NLP model is dependent on a specific input file. This job is supposed to preprocess a given text file to produce this input file for the NLP model (feature engineering). This job will count the occurrences of a word within the given text file (corpus).

There is a dump of the datalake for this under resources/word_count/words.txt with a text file.

---
title: Wordcount Pipeline
---
flowchart LR
  Raw["fa:fa-file words.txt"] -->  J1{{word_count.py}} --> Bronze["fa:fa-file-csv word_count.csv"]

Loading

Input

Simple *.txt file containing text.

Output

A single *.csv file containing data similar to:

"word","count"
"a","3"
"an","5"
...

Run the job

poetry build && poetry run spark-submit \
    --master local \
    --py-files dist/data_transformations-*.whl \
    jobs/word_count.py \
    <INPUT_FILE_PATH> \
    <OUTPUT_PATH>

Citibike

This problem uses data made publicly available by Citibike, a New York based bike share company.

For analytics purposes, the BI department of a hypothetical bike share company would like to present dashboards, displaying the distance each bike was driven. There is a *.csv file that contains historical data of previous bike rides. This input file needs to be processed in multiple steps. There is a pipeline running these jobs.

---
title: Citibike Pipeline
---
flowchart TD
  Raw["fa:fa-file-csv citibike.csv"] -->  J1{{citibike_ingest.py}} --> Bronze["fa:fa-table-columns citibike.parquet"] --> J2{{citibike_distance_calculation.py}} --> Silver["fa:fa-table-columns citibike_distance.parquet"]

Loading

There is a dump of the datalake for this under resources/citibike/citibike.csv with historical data.

Ingest

Reads a *.csv file and transforms it to parquet format. The column names will be sanitized (whitespaces replaced).

Input

Historical bike ride *.csv file:

"tripduration","starttime","stoptime","start station id","start station name","start station latitude",...
364,"2017-07-01 00:00:00","2017-07-01 00:06:05",539,"Metropolitan Ave & Bedford Ave",40.71534825,...
...

Output

*.parquet files containing the same content

"tripduration","starttime","stoptime","start_station_id","start_station_name","start_station_latitude",...
364,"2017-07-01 00:00:00","2017-07-01 00:06:05",539,"Metropolitan Ave & Bedford Ave",40.71534825,...
...

Run the job

poetry build && poetry run spark-submit \
    --master local \
    --py-files dist/data_transformations-*.whl \
    jobs/citibike_ingest.py \
    <INPUT_FILE_PATH> \
    <OUTPUT_PATH>

Distance calculation

This job takes bike trip information and adds the "as the crow flies" distance traveled for each trip. It reads the previously ingested data parquet files.

Hint:

• For distance calculation, consider using Haversine formula as an option.

Input

Historical bike ride *.parquet files

"tripduration",...
364,...
...

Outputs

*.parquet files containing historical data with distance column containing the calculated distance.

"tripduration",...,"distance"
364,...,1.34
...

Run the job

poetry build && poetry run spark-submit \
    --master local \
    --py-files dist/data_transformations-*.whl \
    jobs/citibike_distance_calculation.py \
    <INPUT_PATH> \
    <OUTPUT_PATH>

---

⚠️ do not try to solve the exercises ahead of the interview

---

Reading List

If you are unfamiliar with some of the tools used here, we recommend some resources to get started

• pytest: official
• pyspark: official and especially the DataFrame quickstart