'How do I save a dataframe with a list column of same-columned dataframes to parquet with arrow?
I have a dataframe with roughly 1200 columns, one of which is a list. When using arrow::write_parquet(), all columns but the list column are detected correctly. The list column has many NULL values. The values that exist are dataframes themselves with five character columns (always with the same 5 names) and 1 to many rows. (It's possible if the 'NULL' values cause problems I could put zero-row dataframes in their places.)
Is there a way to specify a schema for the one list column and detect the rest of the columns so that I can save the dataframe to parquet?
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|