Create a schema.sql shortcut file

[ruf-io]

Sometimes I want to just create all the sources, views, matviews right away.

I think it'd be nice to standardize by creating a schema.sql file for each demo that lets you do that.

A couple open questions:

1. Does this create some annoying SQL duplication issues between readme and file?
2. Is there a way to execute a SQL file through the provided docker containerized CLI? (so users dont need psql installed)

[morsapaes]

That's a great idea!

Is there a way to execute a SQL file through the provided docker containerized CLI? (so users dont need psql installed)

You can boot Materialize using a .sql script and mzcli (see this example). So, we should be able to just provide the docker run command that does this if folks want to get everything created for them.

[morsapaes]

Ah, we'll want to test this through, since there might be an issue with this method: https://materializecommunity.slack.com/archives/C015KDVS7EV/p1650629278433099

[morsapaes]

Turns out the issue above was related to misconfigured healthchecks!

[bobbyiliev]

Maybe closing this PR and reworking this to use the new dbt init feature that @dehume worked on?

[dehume]

Maybe closing this PR and reworking this to use the new dbt init feature that @dehume worked on?

If you still want to the demo to use the red panda infrastructure, it won't really line up with the dbt init. The dbt init is much more self contained and just uses load generators.