Python Polars: The Definitive Guide

Chapter 7. Beginning Expressions

The goal of this chapter is to introduce expressions, which are what makes the Polars API so powerful and elegant. This chapter forms the basis for the remaining chapters of Part III, “Express”, where we go into more detail regarding specific expressions and how to use them.

Polars Expressions Versus Regular Expressions

Polars expressions should not be confused with regular expressions. A regular expression, or regex, is a sequence of characters that is used to match text. For example, the regex [Pp](ol|and)ar?s matches both pandas and Polars, but it doesn’t match panda or polaris. A few Polars methods do accept regexes, such as pl.col() for selecting columns, and Expr.str.replace() for replacing values. The interactive website RegExr by Grant Skinner and the book Introducing Regular Expressions by Michael Fitzgerald (O’Reilly) are useful resources for learning more about regexes.

Expressions, in Polars, are reusable building blocks that enable you to perform many data-wrangling tasks, including selecting existing columns, creating new columns, filtering rows on a condition, and calculating aggregations. In short, they pop up everywhere.

Expressions have so much to offer that we’ve split their discussion into three chapters, as pictured in Figure 7-1. In Chapter 13 we cover various methods that are accessible through so-called namespaces (explained in the next section).

Figure 7-1. The many expression methods are organized into three chapters ...