In data science, there is a concept called the 80/20 rule.
Strictly speaking, this is more of an empirical observation than a law: of the resources (time and manpower) consumed in the process of performing data analysis, only 20% go to the actual modeling and analysis work, while 80% of the resources are spent on finding, refining, and reconstructing data.
Data is always lacking, and even the data we do collect is often missing information or not formatted correctly.
The analysis results do not match intuition, probably because the data contains strange values.
Anyone who has ever done real data analysis work will agree with this sentiment.

The example data we cover in most classes is usually 'clean' data that has been preprocessed to reduce trial and error.
This textbook covers the basics of preprocessing, collecting more realistic "dirty" data and refining the collected data to suit the purpose of analysis.
Since covering each element in depth (web scraping, API programming, missing value filling, etc.) would require several books, we introduce the most common and basic techniques and provide appropriate examples so that you can apply them to real data.

This textbook contains a lot of Python example code.
You can download these codes directly from the author's website, but since data processing is more of a practical skill to be learned than a systematic academic discipline, I recommend that you don't just run these codes; you input them yourself, go through trial and error, and develop them into useful tools for yourself.
I recommend that you go through trial and error to develop it into a tool that you can use usefully.