Currently, FileFormatUtils supports different APIs to fetch record keys, min/max. We need a new API to project only selected columns and return an iterator or records.
This could be useful for updating secondary index, when file slice has only base file - #12525 (comment)
JIRA info
Currently, FileFormatUtils supports different APIs to fetch record keys, min/max. We need a new API to project only selected columns and return an iterator or records.
This could be useful for updating secondary index, when file slice has only base file - #12525 (comment)
JIRA info