Yury Selivanov
Yury Selivanov
Given time constraints, on day 1, we can just keep the status quo and return strings. If we go with option (2).
* Use Gel schema annotations to specify full qualified name for a Pydantic JSON model -- this is better than making users overload reflected classes - that composes poorly *...
I've doubts about unpacking JSON by default. The whole point of using pydantic is type safety, maybe nobody wants to unpack automatically to non-pydantic raw python dicts?
Consensus reached: we will not unpack JSON automatically if it doesn't have a schema (`std::lang::py::annotation`). We can add a client config-like thing `client.with_auto_json().query(...)` and then unpack. But not by default.
As well as complex scalars -- arrays, arrays of tuples with nested arrays, etc
@vpetrovykh Can you add a test for enum? @elprans we need to fix values editing - moving this to high pri
Complex scalars still need work, iirc -- there's no validation for them
@elprans says it's working already -- please validate
Python is more expressive than TS here, so I'm wondering if we can pull off this (idiomatic Python casting): ```python std.str('aaa') std.array[std.str](...) ``` or if that's umbigous than this (albeit...
Also, I think Python QB is TS QB have diverged so much that TS can't/shouldn't influence Python QB API much at this point (design mistakes count though!)