Pandas: how to add two dataframe with list element ignoring NaN?

I want to use Pandas add operation to replace defaultdict(list) functional. If an index exists only in one dataframe, the result will be NaN. However, I hope the result is the existing list. For example:

In [3]: x = pd.DataFrame({'fnm': [[t] for t in 'abcdef']}, index=[0, 1, 2, 3, 4, 5])                                                                 In [4]: x                                                                                                                                          Out[4]:      fnm  0  [a]  1  [b]  2  [c]  3  [d]  4  [e]  5  [f]    In [5]: y = pd.DataFrame({'fnm': [[t] for t in 'abcdef']}, index=[0, 2, 3, 4, 5, 6])                                                                 In [6]: y                                                                                                                                          Out[6]:      fnm  0  [a]  2  [b]  3  [c]  4  [d]  5  [e]  6  [f]    In [7]: x + y                                                                                                                                      Out[7]:         fnm  0  [a, a]  1     NaN  2  [c, b]  3  [d, c]  4  [e, d]  5  [f, e]  6     NaN    

I expect the result is

0  [a, a]  1  [b]  2  [c, b]  3  [d, c]  4  [e, d]  5  [f, e]  6  [f]  

https://stackoverflow.com/questions/65838398/pandas-how-to-add-two-dataframe-with-list-element-ignoring-nan January 22, 2021 at 10:07AM