Getting rows from RecordBatch

[kskalski]

I'm looking at the async API for reading parquet files and it offers creating stream of RecordBatch. Right now I'm using parquet::file::serialized_reader::SerializedFileReader for getting iterator over rows of the file, but I'm a bit lost on how to achieve similar access when using async and RecordBatch, which just gives raw access to columns.

Is this kind of conversion into Row objects easy to perform using existing APIs? Should row reader over RecordBatch be added as a new feature to parquet::record module? Or is there going to be parquet::file::async_reader::AsyncFileReader?