Turn data.frames with a JSON column into a tbl_json object

[jeremystan]

From David, a data scientist at socure.me:

"My question is whether it can be used on a data frame with a number of regular columns and one json column. Ideally, I'd like to expand just the json column into its constituent columns, and then bind them back with the rest of the dataframe. Is this possible?"

One way to implement this would be through a as.tbl_json.data.frame function that takes a mandatory second argument which identifies the column of the source data.frame that contains the JSON data.

This is useful for:

• JSON that was added to an existing data.frame (perhaps via multiple API calls)
• Tables exported from relational databases that have JSON columns (e.g., Postgres)

This should also support dplyr::data_frame and dplyr::tbl.df source objects.