Trying to figure out how to transform a k-v string that is inside a column where the k-v string is separated by commas, and could contain different keys. The different keys would then be transformed into their own columns, where missing values would contain nulls.
For example,
pl.DataFrame({ "apple": [1, 2, 3], "data": ["a=b, b=c", "a=y, y=z", "k1=v1, k2=v2"]})would look like:
pl.DataFrame({ "apple": [1, 2, 3], "a": ["b", "y", None], "b": ["c", None, None], "y": [None, "z", None], "k1": [None, None, "v1"], "k2": [None, None, "v2"], "data": ["a=b, b=c", "a=y, y=z", "k1=v1, k2=v2"]})once transformed. Does anyone know what is the most efficient way to do this (perhaps without pre-processing of the data, if possible?)
Post a Comment