Stream: compiler development

Topic: state of roc


view this post on Zulip drew (Jun 21 2024 at 16:52):

I have a high-level question -- what is the current state of roc? It seems that things have been in flux for a while, but with the updated module syntax and the ! change complete, are we at a point where things may slow down and focus will return to stability and working towards a 0.1 release? i ask because there still seem to be a lot of seg faults and tooling issues (100% from lsp, for example) and i'm trying to understand how i should think about the current short term goals for the language. none of this is an issue, i'm just curious.

view this post on Zulip drew (Jun 21 2024 at 16:53):

the core of the language feels like it's at a nice spot to me at this point, but i'm definitely not close to the development

view this post on Zulip Brendan Hansknecht (Jun 22 2024 at 06:43):

I don't really have an answer, but I think a number of larger pieces would be wanted before a 0.1 release:

  1. Finishing the module params change
  2. Effect interpreters (large platform interface change that I think is currently stuck on type system bugs)
  3. Explicit allocators passed in from the platform
  4. glue for at least rust being fully usable
  5. Removal of <-.

Some likely nice to haves:

  1. Fixing some performance sharp edges around refcounting and borrow inference.
  2. Surgical linker fully working on all platforms
  3. Dev backends fully complete and wired up (still missing a few minor pieces)

Not sure which of these are true blockers or if I am missing anything, but definitely a solid bit still in flux. That said, most of these don't really effect the look and feel of roc as a language. Most of it is more around platform dev and completeness.

view this post on Zulip Luke Boswell (Jun 23 2024 at 02:55):

I'd love to track these issues down and make a milestone, I was thinking 0.1.0 or something. Would that be a bad idea?

view this post on Zulip Richard Feldman (Jun 23 2024 at 12:03):

personally I’m thinking about 0.1 as something we do when we think there’s enough benefit to justify it

view this post on Zulip Richard Feldman (Jun 23 2024 at 12:03):

I don’t have a checklist or anything in my mind :big_smile:


Last updated: Jul 06 2025 at 12:14 UTC