dictum VS chinook-database

TL;DR: Github repo | Documentation

I have been working on this project called Dictum. Here's the Github repo and Documentation. Looking for some feedback from BI people.

In literally 30 seconds, you can open the blog post, upload a SQLite file, and then query it in the Splitgraph web console (or via HTTP, or via any Postgres client). Authentication is not required.

Here's the 30 second video [0] and end result [1]. If you don't have a SQLite file handy and want to try it yourself, here's a link to the Chinook Database [2].

[0] Video (30 seconds): https://twitter.com/splitgraph/status/1646612048837591040

[1] Repository: https://www.splitgraph.com/miles/chinook-tables/latest/-/tab...

[2] chinook.sqlite: https://github.com/lerocha/chinook-database/blob/e7e6d5f008e...

I’ll be using a Postgres datasource (the famous Chinook database) for my example, using WunderGraph to make it accessible through JSON-RPC to my Next.js frontend.

If you look at a typical SQL query in analytics, it looks something like this (taken from Chinook sample database):

schemacrawler/chinook-database builds on the earlier project. The project has an automatic build using GitHub Actions, which pulls scripts from lerocha/chinook-database, cleans them up, and converts them all to UTF-8 with consistent line-endings. Then are are repacked and redistributed in various ways - as a Java jar file for use in Java programs or in tests with database in Testcontainers. The SQLite database can be directly download from the project site. But probably the best way to use the package it is use the Docker image from schemacrawler/chinook-database to create the Chinook database on a database of your choice, whether the database server is running on a separate host or within a Docker container itself.

https://github.com/lerocha/chinook-database#:~:text=Chinook%20is%20a%20sample%20database,single%20and%20multiple%20database%20servers.