- ENH: Add ability to extract parts of header metadata via
`TableBuilderReader.read_header_metadata()`
- ENH: Add support for reading files with filters
- Change default API for reading dataframes to `TableBuilderReader.read_table()`
Old API name `TableBuilderReader.read()` has been kept for backwards compatibility
- ENH: Add support to remove total columns from data via
`TableBuilderReader.read_header_metadata()`
- ENH: Add support to strip "Total" rows/ columns via
`TableBuilderReader.read_table(*, drop_totals=...)`, which takes values `"rows", "columns",
- ENH: Tables read with `as_index=True` will have the index `dtype` coerced to `int` if possible.
This is to make `join`/`merge`ing on e.g. SA1 indices more straightforward
-
`TableBuilderReader.read_table(*, drop_totals=...)`, which takes values `"rows", "columns",
"both"` or `None`
- REF: cleanup of internals, switch to pytest for tests, improve internal type annotation coverage