If you know me you know that I love Pokémon. It is my favourite video game series going today. It’s been going on for 21 years now, an within the game there is a lot of data, that clearly sits within a database. And being a database guy, I’ve always wondered how the database would be structured, and how the changes to the game would be incorporated.

Long ago, a guy called Veekun created a copy of this database and posted it on GitHub. The GitHub project supports 3 databases: SQLite, MYSQL and PostGreSQL. And so recently I wondered if it were possible to migrate the database to SQL Server. Mainly because I work in SQL Server, but also because I thought it would be a neat project to try out.

So I spun up a VM and followed the Getting Data wiki article on the Pokedex GitHub page. It did take a few attempts of Googling errors, as I had not setup Python before, or used SetX in quite some time. However I fuddled my way through it!

And the result is an SSDT-based project, that currently is failing to compile solely because of the non-existent “BOOLEAN” data type. There are many other issues that I need to fix, like no names for Keys, and plenty of keywords used within the database. So a lot of refactoring. Not to mention getting the data loading scripts to work.

Nevertheless, it’s a starting point, and certainly I will blog about the progress I make on refactoring the database, the limitations of SSDT and the workarounds.

I’m pretty excited about this project and hope it will grow into something that actually sees use as opposed to just an academic experiment.