Python API

datui.view() opens a file, a URL or a Polars frame in the terminal, and can hand the final view back. Use datui from Python is the guide.

import polars as pl
import datui

df = pl.DataFrame({"city": ["Oslo", "Lima", "Pune"], "temp_c": [4.5, 19.0, 27.5]})
result = datui.view(df, capture=True, row_numbers=True)
if result is not None:
    print(result.collect())

datui.view

The signature; data is the one argument it needs:

datui.view(data, *, capture=False, options=None, **kwargs) -> polars.LazyFrame | None

A path or URL is read as the command line reads it (s3://, gs://, abfss://, http(s)://, globs), and the open’s options apply. A frame is handed over as its serialized plan, and only display options apply to it. An az://container/path URL takes its account from the Azure environment variables or from the config’s one Azure connection.

Return value

None, unless capture=True and a dataset was open at quit. Then a polars.LazyFrame: the applied query, filters, sort, drill-down, reshape and column order, over every matching row. It is a plan, not the rows datui showed: collecting it runs the plan again with Python’s Polars, rereading files that must still exist.

Errors

datui.DatuiOptions

The same options as keywords, made once and passed as options=. A value is checked when it is made, with the errors above.

import datui

with open("readings.csv", "w") as f:
    f.write("# exported 2024-03-01\nsensor;value\na;1.5\nb;2.0\n")
opts = datui.DatuiOptions(delimiter=";", comment="#")
datui.view("readings.csv", options=opts)

A value is written as the flag or key takes it: True or False, a number, a string, a list of strings, or a pathlib.Path. delimiter also takes the character’s code (ord(";")).

Options