• The table.lookup() method now has an optional second argument which can be used to populate columns only the first time the record is created, see Working with lookup tables. (#339)
• sqlite-utils memory now has a --flatten option for flattening nested JSON objects into separate columns, consistent with sqlite-utils insert. (#332)
• table.create_index(..., find_unique_name=True) parameter, which finds an available name for the created index even if the default name has already been taken. This means that index-foreign-keys will work even if one of the indexes it tries to create clashes with an existing index name. (#335)
• Added py.typed to the module, so mypy should now correctly pick up the type annotations. Thanks, Andreas Longo. (#331)
• Now depends on python-dateutil instead of depending on dateutils. Thanks, Denys Pavlov. (#324)
• table.create() (see Explicitly creating a table) now handles dict, list and tuple types, mapping them to TEXT columns in SQLite so that they can be stored encoded as JSON. (#338)
• Inserted data with square braces in the column names (for example a CSV file containing a item[price]) column now have the braces converted to underscores: item_price_. Previously such columns would be rejected with an error. (#329)
• Now also tested against Python 3.10. (#330)