Add pl.Config.show_full (or something similar)

[mcrumiller]

Description

Sometimes you may want to simply print an entire table, and see everything, with no truncation anywhere. Sometimes you may want the string representation of the full table, no matter how big it is.

Right now, we can manually set the fmt_str_lengths to 999999 and tbl_rows/tbl_cols to 99999, but I always have to go back to the API to look up what the actual names are, since some words like "fmt" and "str" are abbreviated but others like "lengths" are not. It would be very nice to have a single configuration option to print the full table.

This does open the door for floating point questions and really long lists, so some feedback on this idea would be welcome.

[cmdlineluser]

Just on the point of printing larger lists, it reminds me of the request for an equivalent to pandas display.expand_frame_repr option: #7665

import polars as pl
import pandas as pd

df = pl.DataFrame({
    "A": [ list(range(20)) for _ in range(5) ],
    "B": range(5),
    "C": [ "foo" * 20 for _ in range(5) ]
})

with pd.option_context(
    "display.expand_frame_repr", True,
    "display.max_columns", None,
    "display.max_rows", None,
    "display.max_colwidth", None,
): 
    print(df.to_pandas())

                                                                        A  B  \
0  [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]  0   
1  [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]  1   
2  [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]  2   
3  [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]  3   
4  [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]  4   

                                                              C  
0  foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo  
1  foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo  
2  foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo  
3  foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo  
4  foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo  

---

.glimpse() was previously suggested to me - which is useful, but (for me) can be a bit harder to read for longer data:

df.glimpse(max_items_per_column=-1) 

Rows: 5
Columns: 3
$ A <list[i64]> [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19], [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19], [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19], [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]
$ B       <i64> 0, 1, 2, 3
$ C       <str> 'foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo', 'foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo', 'foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo', 'foofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoofoo'

[evbo]

It would be awesome if both Rust and Python had a way of simply printing everything.

I'd take df.glimpse(GlimpseArgs{expand_all: true, ..Default::default()}) any day

Even if it's extremely poor performance - I think it goes without saying this feature is only for desperate debugging