Hi all - looking for insight into the maturity/capability of Pulumi's Snowflake provider as we setup DataOps infrastructure in our Snowflake environment. • We'll use either Terraform or Pulumi for "infrastructure" objects (warehouses, databases, roles, etc.) • We'll likely use dbt for transformations (2 to 3 months away) • Our biggest need right is a DataOps capability for our ingestion layer, if possible ◦ i.e. some kind of versioning and dependency management for things like file formats, stages, and table DDLs in our ingest layer ◦ ideally sophisticated support for change management ▪︎ one example: column changes such as name or data type changes should run

ALTER TABLE ... ALTER COLUMN;

... and not

ALTER TABLE ... DROP COLUM ...; ALTER TABLE ... ADD COLUMN ...;

The last example is how Snowflake's Terraform provider handles column changes, which is not awesome. And I see that Pulumi's Snowflake provider is based on Snowflake's Terraform provider. So, my bigger question is - has Pulumi built the provider somehow more robustly than the underlying Terraform provider? More specifically, Snowflake still has many database object resources such as

snowflake_table

in "Preview" status with a big red warning in the docs, but I see no such statement regarding the snowflake.Table resource in the Pulumi docs. So, does it have some level of "enterprise" capability/support above and beyond that of Snowflake's (quite limited and currently unsupported) Terraform provider? Thanks!