Summary

In this chapter, we saw how DuckDB offers a range of data types, functions, and SQL enhancements that make it a versatile data processing tool for working with semi-structured data. We covered some of DuckDB’s key nested data types – LIST, MAP, and STRUCT – which we then built on so that we could explore working with semi-structured data. This included working with DuckDB’s LIST data type and seeing the flexibility it offers, as well as covering a selection of list processing functions and SQL enhancements that enable some powerful and effective list processing patterns. We then turned to look at DuckDB’s json extension and learned how we can create JSON objects from existing DuckDB data, import JSON data into DuckDB, and work with semi-structured data produced from JSON imports. We also saw how DuckDB enables us to work effectively with JSON data containing inconsistent schemas. We concluded this chapter with a set of practical examples that...