Presented by

Abstract

JSON is the data interchange format of choice. But it is problematic in relational databases like PostgreSQL or MySQL because it violates so many basic assumptions. But all the databases support storing JSON data. But you cannot use many SQL functions, such as aggregates or simple comparisons, with JSON data. JSON_TABLE() takes your JSON-formatted data and temporarily converts it into a relational format for use with traditional SQL functionality. We will start with a simple case, build to more complex cases with embedded JSON and missing data, and then proceed to error handling.

There are good reasons to store data in your database in JSON. And you can mitigate the violations of Codd's rules by using JSON_TABLE(), living happily ever after.