Hacker News Reader

Comment by darthrupert | Hacker News Reader

Hardly ever is "good REPL" one of the features of these new languages. Do they think it's irrelevant or is making such a thing too difficult compared to all these more CSy features?

View on HN

REPLs are often difficult to reconcile with AoT compilation and static analysis, particularly in the context of things like type inference (at least, in a way that preserves semantics). It's on my mental todo list, but not a priority for me.

How fast is the compiler? In many cases a report can just be syntactic sugar for compiling and running an accumulating log of source code lines. It’s not really important how it works under the hood as long as it’s mostly transparent to the user.

As a side effect, it might also be a good way to keep your compile times down.

Fast enough for this approach to work for small examples. For longer examples, I'm less confident: I've yet to implement a proper module system (other than a slightly more principled version of C's #include) so this currently means compiling the entire standard library on every REPL line. Thankfully, the compiler can still handle this within about 100 milliseconds on my machine. Perhaps when I have a more rugged module system I can look into this more, thanks for the suggestion!

What if one of those lines fires the missiles?

That's not even close to being adequate for the kinds of things good REPLs are good for.

Do you suppose that might be an outmoded historical bias? I'm wondering if maybe the advancement of hardware has made it so that the costs of AOT compilation and a type system should no longer stand in the way of delivering a reasonable REPL experience. Swift's "playgrounds" seems to do alright — a real outlier in this regard.

Haskell has a REPL. No idea whether it’s considered good though.

Reply on news.ycombinator.com