Collapse rows into a single row in polars

str.join

Polars allows you to collapse multiple rows into a single row, combining the values from all the rows into one. Below is a dataframe showing parts of a phone number.

shape: (3, 2)

Represent	Value
str	i64
"Area code"	563
"Exchange code"	384
"Line number"	9905

Collapse rows into one

We can construct the full phone number from the three rows using the Polars expression str.join like this:

(df
 .select((pl.col('Value').str.join('-')).alias('Phone_No'))
 )

shape: (1, 1)

Phone_No
str
"563-384-9905"

Note

This is different from the implode function, which returns a list of values instead of a string.

(df
 .select(pl.col('Value').implode().alias('Phone_No'))
 )

shape: (1, 1)

Phone_No
list[i64]
[563, 384, 9905]

Check out my Polars course to learn more.