resolve_column_selection

resolve_column_selection(
    data,
    selectors,
    *,
    allow_none=True,
    require_match=True,
    unique=True,
)

Resolve flexible column selectors into a concrete list of column names.

Parameters

Name	Type	Description	Default
data	DataFrame or GroupBy	Data source used to validate column existence.	required
selectors	various	- str: treated as a literal column name. - Sequence[str]: a collection of literal column names. - re.Pattern: matches columns via pattern.search. - Callable: receives the column Index and must return an iterable of column names. - None: permitted when allow_none is True (default), yielding [].	required
allow_none	bool	Allow None selectors without raising an error. Defaults to True.	True
require_match	bool	Raise ValueError when a selector does not match any columns. Defaults to True.	True
unique	bool	Return deduplicated column names while preserving order. Defaults to True.	True

Returns

Name	Type	Description
	list[str]	Ordered list of columns satisfying the selector(s).

Examples

import pandas as pd
import pytimetk as tk
from pytimetk.utils.selection import contains

df = pd.DataFrame({"grp": [1, 1], "value": [10, 20], "category": ["a", "b"]})
tk.resolve_column_selection(df, ["value", contains("cat")])

['value', 'category']

import polars as pl
import pytimetk as tk

pl_df = pl.DataFrame({"grp": [1, 1], "value": [10, 20], "extra": [0.1, 0.2]})
tk.resolve_column_selection(pl_df, ["value"])

['value']