It's a bit clunky using varchar(max) for these types, but it solves the issue without needing to know what the maximum possible length of each text field is and having a special dispatch which takes the length (that other databases don't care about).
And imo it's better to just add this support and then someone can optimise this later if they have a problem with varchar(max).
Overview
Adding Redshift support - seems like with the new changes this is fairly trivial. The only issue I had running the test suite was the one described in https://github.com/brooklyn-data/dbt_artifacts/issues/60.
Which is trivially solvable (not necessarily in a nice way) by adding redshift specific dispatches for string type helpers that return a varchar(max).
Update type - breaking / non-breaking
What does this solve?
https://github.com/brooklyn-data/dbt_artifacts/issues/60 Adding super basic redshift support as per issue 60.
Outstanding questions
It's a bit clunky using varchar(max) for these types, but it solves the issue without needing to know what the maximum possible length of each text field is and having a special dispatch which takes the length (that other databases don't care about).
And imo it's better to just add this support and then someone can optimise this later if they have a problem with varchar(max).
What databases have you tested with?