Is there an efficient way to aggregate data from a jsonb column in postgresql? Given the table

userdata(id STRING, data JSONB)

and the data

123.abc, { "dis": ["close"] }
123.abc, { "purpose": {"score": 0.1, "text": "hi"}, "dis": ["hello", "close"]}
123.abc, { "dis": ["bye"], "dir": 1}
123.abc, {}
567.bhy, { "dis", ["close"]}

I’m trying to get the following result:

123.abc, { "dis": ["close", "hello", "bye"], "purpose": {"score": 0.1, "text": "hi", "dir": 1} 
567.bhy, {"dis", ["close"]}

As you notice, I want unique values of keys (for e.g. "dis") across all records indexed by the id.

I tried getting values using jsonb_agg and jsonb_array_elements. I could not aggregate all keys and distinct values. I couldn’t figure out how to use jsonb_each to get all keys

What I tried is something like this to get one key.

select id,
       (select jsonb_agg(t->>'dis') from jsonb_array_elements(data::jsonb) as x(t) where t->>'dis' is not null) as sdata
from mytable where id='123.abc'

Any help with the query is appreciated.