There are information about the file that could be relevant to the user.
Global information
This could be added in a row above the table.
- Number of rows/columns (not immediately available without scrolling)
- Size of file
Column level information
This could be added under the column name, or on hover
- Type of the column data type (string, int, categorical... nullable?)
- Easily available statistics, for instance for Parquet we may get for free min, max, null count, sometimes distinct count
- Optionally, we could compute other statistics (more expensive) on demand: mean/media/stddev distribution...
There are information about the file that could be relevant to the user.
Global information
This could be added in a row above the table.
Column level information
This could be added under the column name, or on hover