Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

Hello! I just watched <https://www.youtube.com/watch?v=H6Wslbp7jjk|this> and I was wondering: when is it better to use a Variable in place of a Block (like in this exemple, I now use a String Block to store the environment)?

Hi Adrien!  This is a good question - you can think of variables sort of like environment variables; they are always strings (so don't have an inherent type other than String) and are typically single valued.  Blocks, on the other hand, are generally typed with rich schemas.

If you are using a string block you can continue to do so, but I highly recommend migrating that to a Variable instead - I think it will be less heavy than a block for you

Thanks a lot for your answer! I understand, but in the meantime, the process to setup blocks and variables are quite the same (you need to have a process which set them in order for the flows to be able to use them).

So the goal would have been to use blocks when you have to use 3rd party libraries (like dbt_prefect) and `variables` when you need to define variables related to your own environment?

Yup I think that's a really good way of thinking about it!