Maximilian Roos

> It is my hope that existing PRQL queries will work forever This is the goal of the versioning system! From the book: https://prql-lang.org/book/queries/dialect_and_version.html > The compiler will compile for...

Another advantage of implementing this — we could release experimental features without worrying about backward compatibility or doc warnings. For example, the we're still figuring out exactly `switch` / `match`...

> As I read the statement "... The compiler will compile for the major version of the query..." my fear is that this rule would permit a 2.0 version to...

Great, agree with your points re attempting to anticipate future breaking changes. Look forward to you joining. (to set expectations — we're probably not going to discuss this specific issue...

I broadly agree. I do find "string" a bit engineer-y — but maybe "inline" isn't that much better. I'm +1 on `from_string` unless we can think of a friendlier alternative

> ```elm > > select (round precision 0.12345) > --------- > \ first arg of function round expected int, found column > ``` This would indeed be great! --- >...

> But I do think that applying a function to a column should keep the column name. Let's call this "variable name propagation". So these two would be equivalent: >...

> This feels like the kind of magic that comes around to bite you in the ass. rofl, my thoughts exactly I'll close then? No permanent verdict obv.

I remembered we discussed whether we could use Chumsky "gradually" on a dev call a couple of weeks ago — i.e. Pest at the front-end and Chumsky consuming Pest `Pair`s...

Yes!