Ritchie Vink

> Another simpler solution for now is that I could introduce a new parameter output_format, where you could specify for now "pandas" (the default) or "dict". With dict the functions...

This is already great! We can improve our `from_dict` performance a lot. Until now, it was not really use for heavy load data injection in polars, so we didn't gave...

It seems that the `DataFrame` has an `Object` dtype. This is a last resort data type for anything we don't know. Different from pandas, object types are not string types...

Awesome! Thanks a lot. I also improved `from_dict` performance. Once it is released wrap this in polars. 👍

> So you would create a method in polars to read spss that wraps pyreadstat.read_sav? Yes. :)

> I guess you are asking specifically if this can be done without copying the values into a NumPy array and then back into the `Vec`? Indeed. :) Yes all...

> Is there any progress on this case? > > I'm building full-stack scientific applications where a precise data model is very helpful, and I would like to try combining...

> Is there a fundamental reason why polars do not support dot-notation? Yes! Ambiguity. Dot notation implies attributes. Attributes should be static. With dot notation it can be attributes or...

Could you make a PR for this?

In latest release predicate pushdown is sent to pyarrow, so `scan_ds` should be partition aware if pyarrow is.