Long to wide, filling missing element with `nil`

[KitaitiMakoto]

Hello,

Can I reshape a long data frame to a wide data frame filling missing element with nil?

When I have long data frame such as:

df = DataFrame.new(
  names: %w[name1 name1 name1 name2 name2 name3 name3 name3],
  NAME: %w[One Two Three One Three One Two Three],
  VALUE: [1.1, 2.1, 3.1, 1.2, 3.2, 1.3, 2.3, 3.3]
)

I want wide data frame as:

wide = DataFrame.new(
  names: %w[name1 name2 name3],
  One: [1.1, 1.2, 1.3],
  Two: [2.1, nil, 2.3],
  Three: [3.1, 3.2, 3.3]
)

How can I do this? to_wide raises error: #to_wide with missing values(DataFrameReshapingTest::to_wide): IndexError: element size differs (3 should be 4)

[heronshoes]

Thank you @KitaitiMakoto .

I think this feature is useful.
I think it is better to separate steps in filling missing records.

1. Create Cartesian Product(直積) of df.names.uniq and df.NAME.uniq.
2. Fill missing records in df.
3. #to_wide

Step 1 may require new method to create DataFrame from two Vectors.
Step 2 may use #join ? (I'm not sure now).

I will code them and compare performance with your proposal.