Feedback on SPEC.md #222
Replies: 3 comments 2 replies
-
Thx! I'll update the spec accordingly. I'm pressure testing the spec against a Python implementation right now, so it might take me a bit to push the changes. |
Beta Was this translation helpful? Give feedback.
-
I can see why this is confusing. The attribute types are NOT the same as Recap types. The reason I used All that said, I will clarify the spec to make this more obvious.
So, the That said, I suspect the better way to model a list is with the following attributes:
Thus, a LinkedList (with unlimited length) would have I didn't include the
Yea, this is a tricky one. So, Recap's type system doesn't have literals. But again, similar to the first comment, the attribute type for
Yep. I'll add this. |
Beta Was this translation helpful? Give feedback.
-
I've updated the spec accordingly: https://github.com/recap-cloud/recap/blob/main/SPEC.md |
Beta Was this translation helpful? Give feedback.
-
Hey @criccomini, a few random thoughts:
int32
before actually introducing this type. An explainer upon the first reference would be good ("short-form alias for an int with 32 bits, see ... for details")bits
, how would alist[string(bytes=100]
be defined? in contrast, no attributes of that kind are mentioned for map keys/values; alsosigned
is mentioned for list in the example, but not the definition; I think this is mostly down to how types are represented in definitions and examples consistentlyBeta Was this translation helpful? Give feedback.
All reactions