[Python] Creating Python Polars DataFrame/LazyFrame from Dict representing single Frame row?

I want to create a Polars DataFrame or LazyFrame (I'm not sure what is appropriate) from a list of dictionaries where each of the dicts represents one row of data. Each individual dict has fields of various data types and most importantly some fields have array-like data. The array-like data related to on key can vary in length from dict to dict. A single row dictionary looks like this:

{
'col1': int,
'col2': int,
'col3': str,
...
'colA': str:timestamp,
'colB': int(long),
'colC': float,
...
'colK': bool,
...
'colL': str[L], <-- length differs from dict to dict
'colM': float[M], <-- length differs from dict to dict
'colN': int[N], <-- length differs from dict to dict
...
}


Currently, I am trying to do the following:

for row_dict in list_of_row_dicts:
dataframe = pd.DataFrame(row_dict)
slice_dataframe = slice_dataframe.join(other=dataframe)


This fails with the error message:

polars.exceptions.ShapeError: could not create a new DataFrame: height of column 'colL' (x) does not match height of column 'colM' (y)
​

How can I achieve what I am trying to do?

Continue reading...