liquibase VS schemachange

While I respect the work, Liquibase's code base is quite messy... https://github.com/liquibase/liquibase

Have you looked at liquibase?

For those in search of a open source solution with commercial support have a look at https://www.liquibase.org/

I haven't used it for a number of years but it sounds like something like liquibase (https://www.liquibase.org/) might be what you're looking for.

Liquibase

We use https://github.com/liquibase/liquibase, as most of Dev are used to and comfortable with it.

Highly recommend you use either Liquibase or Flyway for schema management, and Testcontainers pg to run a transitive database for your tests.

I haven't used this so I'm not sure it does exactly what you're looking for but I had a DBA recommend it to me as a possible tool to start using as we overhaul our pipelines. https://www.liquibase.org/

Our organization is transitioning from Microsoft SQL Server to Snowflake. We could use SQL Server Database Projects to deploy schema changes to our data warehouse within the SQL Server space. With Snowflake, the default tool is schemachange. Does anyone know of a database DevOps tool for Snowflake that is more like SQL Server Database Projects?

We use schemachange to deploy scripts, and we have a database prefix variable that we pass in to denote which set of clones this is, so you can spin up your own clone - so any scripts that work across databases have use that in jinja. We have a CICD pipeline in Jenkins (but I also have this replicated in a local Jupyter notebook) where we clone the DBs with this same prefix, apply the schemachange scripts from our stacks (we have multiple stacks and a deployment order through them to manage cross dependencies) and then you're good to go.

We have tried Terraform and others like Snowchange and in the end, we use Snowchange: - Terraform for Snowflake has a very tricky thing is that the table is replaced whenever Terraform detects a change. This is OK for AWS resources but it is bad for data thing like Snowflake - you don't want any change, unless you specifically want to. - We then have to use the second one, Snowchange. This one idea is simple: it marks the SQL files with versions, such as V0.1.1. Snowchange will only run the files if and only if the files are bigger than the biggest recorded in a table in Snowflake (which is done by itself). For example, if the largest version is V0.1.1, and some user accidentally changes the file of V0.1.0, since it is smaller than 0.1.1, the changes are not deployed. Users have to manually the file name to 0.1.2 for it to run.

Schemachange https://github.com/Snowflake-Labs/schemachange

I've trialled schemachange and liquibase which are change script based tools. I've ruled out a whole load of other tools that are either change script based tools or don't support Snowflake, including the following:

For Snowflake, we ended up using this script: https://github.com/Snowflake-Labs/schemachange