What the difference between running: 1.

meltano invoke --dump=catalog tap-salesforce

2.

meltano invoke tap-salesforce --discover

#2 seems to call out to salesforce every time, but #1 seems to cache the catalog somehow. So it would seem that if I want to run this in prod #2 is the right command but I wanted to make sure I understood the difference.

The most recent GitHub issue that collates all the previous issues about cached catalog should be this one: https://github.com/meltano/meltano/issues/7899

Ok cool. So from what I can tell the difference is that

dump

tries to use the

cache

, but

--discover

doesn't. Also, if iI don't want to ever use the cache (with

dump

) I should just delete the

.meltano/run/<tap>

folder and it will re-discover the catalog. We spin up new containers for each run so I'm guessing this won't be an issue either way, but it's nice to understand what's going on.