@huangkuan this is great feedback - thanks!
For folks watching this, some context: @huangkuan is a The Graph Foundation Wave 2 Grantee who has been working on Bubbletea
At the moment, Bubbletea’s python lib is bypassing this “limitation” by invoking multiple requests, aggregating and stitching the results, so the user doesn’t have to. We have ideas that will enable more of this Data Science type of use cases (among others, revamping GraphQL API), so community feedback is important.
I believe the current limitation in place is set purely to try to avoid bad performance, but maybe, in the future, we could allow users to bypass the default behavior.