Delete rows with matching unique key

[matthewrj]

Description & motivation

This introduces "merge" behaviour for insert by period to match the standard incremental materialisation. This allows insert by period to be used with models where rows change over time.

I simply copied the code from the standard incremental materialisation here: https://github.com/dbt-labs/dbt-core/blob/8442fb66a591c4f5353f8baaaaf62f846c5f5171/core/dbt/include/global_project/macros/materializations/models/incremental/merge.sql#L50.

Checklist

• I have verified that these changes work locally on the following warehouses (Note: it's okay if you do not have access to all warehouses, this helps us understand what has been covered)
  • BigQuery
  • Postgres
  • Redshift
  • Snowflake
• I followed guidelines to ensure that my changes will work on "non-core" adapters by:
  • dispatching any new macro(s) so non-core adapters can also use them (e.g. the star() source)
  • using the limit_zero() macro in place of the literal string: limit 0
  • using dbt_utils.type_* macros instead of explicit datatypes (e.g. dbt_utils.type_timestamp() instead of TIMESTAMP
• I have updated the README.md (if applicable)
• I have added tests & descriptions to my models (and macros if applicable)
• I have added an entry to CHANGELOG.md

[joellabes]

@matthewrj Fancy! Thanks for opening this. It'd be nice to have this process support deletes 👍

Two conflicting thoughts here:

1. I'd love to see some tests around this to ensure that it behaves correctly. I think this would be a series of seeds with different stages of progress, but I don't know how to cause it to re-run without multiple invocations of the automated tests, which feels a bit fragile...
2. We're still thinking about the future of this macro (see Modernize legacy insert by period materialization #410), so I don't want to ask you to do a bunch of work that might go unmerged for a while.

For now, please don't be offended if this sits on ice for a bit - we're about to dive into the future of dbt-utils with a vengeance, so I expect there to be some progress in the next month or two.

[joellabes]

Hi @matthewrj, an update on this - have you seen #487? In short, this materialization will move into the experimental features repo in the next month or so. When that happens, we'll come back to this PR!