Anatoly Bugakov

Nooby question: building a code which uses f64s to wasm32-wasi - Would that lead to errors/unexpected behaviour or is that fine?

> I should note: the above code _replaces_ the column with new values, but the same approach using a left join should work for mapping values for new columns. Thank...

@ritchie46 , do you think it's possible to conver LazyFrame from Python to Rust and back like you did [here ](https://github.com/pola-rs/polars/tree/master/examples/python_rust_compiled_function) with Eager frame?

> I have a setup of a crate that does this for you hidden behind pyo3 bindings. But haven't yet had the bandwidth/priority to finish this. @ritchie46 that would be...

2. `let mut ac = self.inputs[0].evaluate_on_groups(df, groups, state)?;` of `polars\polars-lazy\src\physical_plan\expressions\apply.rs` , **here input is apply with returns_scalar = True and therefore is AggState::AggregatedFlat**. As such, on this line this code...

@ritchie46 1. Is harder. ` let len = iters[0].size_hint().0;` returns 0 and therefore this code gets executed, which seems to be incorrect (correct me if I am wrong): ```rust if...

Thanks @ritchie46 , > Are you sure the `aggregated_flat` is correct? I don't really understand how that return value is valid. I am doubting if we should have `apply_multiple` as...

@ritchie46 , no worries, it's Saturday :) > On mobile now. But shouldn't aggregated_flat not return a length equal to the groups? > Currently it returns a single valued series,...

Many thanks, much appreciated! :)

Also, if I do ``` let res = df.lazy().with_column( col("foo").arr().max().alias("foo1") ); dbg!(res.collect()); ``` I get: ``` ShapeMisMatch( Owned( "Could not add column. The Series length 3 differs from the DataFrame...