cudf.core.column.string.StringMethods.filter_characters#

StringMethods.filter_characters(table: dict, keep: bool = True, repl: str | None = None) → SeriesOrIndex[source]#

Remove characters from each string using the character ranges in the given mapping table.

Parameters:

tabledict: This table is a range of Unicode ordinals to filter. The minimum value is the key and the maximum value is the value. You can use str.maketrans() as a helper function for making the filter table. Overlapping ranges will cause undefined results. Range values are inclusive.
keepboolean: If False, the character ranges in the table are removed. If True, the character ranges not in the table are removed. Default is True.
replstr: Optional replacement string to use in place of removed characters.

Returns:

Series or Index.

Examples

>>> import cudf
>>> data = ['aeiou', 'AEIOU', '0123456789']
>>> s = cudf.Series(data)
>>> s.str.filter_characters({'a':'l', 'M':'Z', '4':'6'})
0    aei
1     OU
2    456
dtype: object
>>> s.str.filter_characters({'a':'l', 'M':'Z', '4':'6'}, False, "_")
0         ___ou
1         AEI__
2    0123___789
dtype: object