[FIXED] Polars - get hour from Time column

Issue

I have a Polars df with a column 'Checkout Time' in a string format, e.g '17:54'.

I'm converting to a time format with: df = df.with_columns(pl.col('Checkout Time').str.to_time('%H:%M'))

How would I then get the actual hour of the time as an integer?

I could do it with map_rows but is there a better way than this?

df.with_columns(
    df.select('Checkout Time').map_rows(lambda x: x[0].hour)
).rename({'map': 'Hour'})

But this seems annoying, is there a way to do it directly in with_columns?

Some example data:

data = np.array(['06:30', '17:45', '18:32'])
df = pl.from_numpy(data, schema=["Checkout Time"])
df = df.with_columns(pl.col('Checkout Time').str.to_time('%H:%M'))

Solution

You can use polars.Expr.dt.hour directly:

import polars as pl
import numpy as np


data = np.array(['06:30', '17:45', '18:32'])
df = pl.from_numpy(data, schema=["Checkout Time"])
df = df.with_columns(pl.col('Checkout Time').str.to_time('%H:%M'))

df = df.with_columns(Hour=pl.col('Checkout Time').dt.hour())

>>> df
shape: (3, 2)
┌───────────────┬──────┐
│ Checkout Time ┆ Hour │
│ ---           ┆ ---  │
│ time          ┆ i8   │
╞═══════════════╪══════╡
│ 06:30:00      ┆ 6    │
│ 17:45:00      ┆ 17   │
│ 18:32:00      ┆ 18   │
└───────────────┴──────┘

Answered By - FJSevilla