Polars / Python Limits Number of Printed Table Output Rows

Question

Does anyone know why polars (or maybe my pycharm setup or python debugger) limits the number of rows in the output? This drives me nuts.

Here is the polars code i am running but I do suspect its not polars specific as there isnt much out there on google (and chatgpt said its info is too old haha).

import polars as pl

df = pl.scan_parquet('/path/to/file.parquet')
result_df =(
        df
        .filter(pl.col("condition_category") == 'unknown')
        .groupby("type")
        .agg(
            [
                pl.col("type").count().alias("counts"),
            ]
        )
    ).collect()
print(result_df)

Embrey

It's a string column, so hopefully set_fmt_str_lengths works? – Accretion

Tried and didnt work: with pl.Config() as cfg: cfg.set_fmt_str_lengths(200000) print(result_df) – Embrey

Answer 1

Looks like the following will work. Thanks to @wayoshi for sharing this. I will say that the defaults are way too conservative!

with pl.Config(tbl_rows=1000):
    print(result_df)

or throw this at the top of your script if you prefer to not manage contexts.

import polars as pl

# Configure Polars 
cfg = pl.Config()
cfg.set_tbl_rows(2000)

Embrey

Oh I misread the question as string truncation and not row, my bad. – Accretion

@Accretion Me too. I edited the question to be more explicit/direct. – Paramorphism

Answer 2

You can just put this at the top.

import polars as pl
pl.Config(tbl_rows=50)

You can do 50 rows and 10 columns like this:-

import polars as pl
pl.Config(tbl_rows=50, tbl_cols=10)

Selfexistent