It's a GREAT pattern. Parse + dump 70GB in the bucket (8...
@simonw @magnetikonline It's a GREAT pattern. Parse + dump 70GB in the bucket (8 hrs). Walk each doc, using json_extract to make a table (20 min). Use json_[each|tree] to make one-to-many relations (like the "to" fields of an email). Don't try to make "real" primary keys, just index liberally.
➛