Orson Peters

We already use `.len()` all over the place: https://docs.pola.rs/py-polars/html/search.html?q=len.

> Also, I think a lot of the SQL users will be sad that pl.count() changes behaviour. The SQL front-end would not change due to this. `COUNT(*)` would still work...

> min(null,1) should not be 1. @arnabanimesh In SQL you have `LEAST` and `GREATEST` functions, and they do in fact behave as `LEAST(null, 1) == 1`. --- Our consistent principle...

> It seems it is not the case in mysql atleast: [Link](https://www.w3resource.com/mysql/comparision-functions-and-operators/least.php#:~:text=LEAST()%20returns%20NULL%20if%20any%20of%20the%20input%20values%20are%20NULL.) It is in Postgres and DuckDB. It does appear to have been standardized to propagate nulls in SQL...

> In Python, we often parse None input as 'input was not specified'. It is definitely not always the same as lit(None) which is an expression of data type Null....

It's not exactly clear to me what you're asking. Could you make your example a bit more minimal?

Could you please edit your post to have actual code / output instead of screenshots?

This should still support negative arguments as we discussed.

@mcrumiller I understand, I commented more in case someone else wanted to review/merge in the current state.

I think this implementation is incorrect when the scale and precision are not equal.