Changes in version 0.0.4 (2023-02-20)                  

  - Replaced now-defunct dplyr::select_vars calls with
    tidyselect::vars_select
  - Support (optionally) using listviewer::reactjson to view schemas

                 Changes in version 0.0.3 (2018-11-14)                  

  - Initial release
  - To support read schemas, array_type, binary_type, boolean_type,
    byte_type, character_type, date_type, double_type, float_type,
    integer_type, long_type, map_type, numeric_type, string_type,
    struct_field, struct_type, and timestamp_type are provided. These
    will return java references needed for schema definition.
  - sdf_select enables access to nested fields (within maps, structs,
    and arrays of structs).
  - sdf_explode will convert arrays of structs to simple structs by
    duplicating rows over each unique array entry. Also works on map
    types if is_map=TRUE.
  - sdf_nest and sdf_unnest are similar to the similarly named tidyr
    functions. sdf_unnest essentially chains sdf_explode and sdf_select
    with some schema inspection.
  - sdf_schema_json and sdf_schema_viewer support schema interrogation