Hi, I'm using meltano to load some data to Postgres. Wandering how I should handle unraveling nested data? My table has columns such as

customer_id, start_date, plan

, where the 'plan' data looks like:

{
  "id": "price_abc",
  "active": true,
  "amount": 999,
  "object": "plan",
  "created": "2022-09-09",
  "product": "prod_abcd",
  "currency": "usd"
}

and another tap-associated 'metadata' col has data like

{emails: <mailto:inigo@montoya.com|inigo@montoya.com>

. ...I changed the postgres

data_flattening_max_level

to 1, which flattened out 'Plan' attributes, but not the 'metadata' attributes. Some questions: •

meltano select tap-stripe --list --all

shows the attribute subscriptions.metadata, but not s_ubscriptions.metadata.email_. Does this mean i'd have to handle subscriptions.metadata unraveling later through DBT/ in transformation? • Is it better practice to explicitly pick and choose specific grandchild streams(?), like _subscriptions.plan.amoun_t, subsciptions.plan.id, or to increase the flattening level?

How deep is your metadata column? If you know the depth, you should simply be able to change the data_flattening_max_level=YOUR_DEPTH (e.g. 3) and the columns should be created automatically. - Or is it an issue with the quotes that just one type is flattened? In general you could also use a mapper to achieve this, in case you want to do more things before loading. (https://docs.meltano.com/concepts/plugins#mappers)