Rewrite AST prior go type checking to avoid inter-realm type check errors. #4156
Conversation
github-project-automation
bot
added this to
🧙♂️gno.land core team and 💪 Bounties & Worx
🛠 PR Checks SummaryAll Automated Checks passed. ✅ Manual Checks (for Reviewers):
Read More🤖 This bot helps streamline PR reviews by verifying automated checks and providing guidance for contributors and reviewers. ✅ Automated Checks (for Contributors):No automated checks match this pull request. ☑️ Contributor Actions:
☑️ Reviewer Actions:
📚 Resources:Debug
|
gnovm/pkg/gnolang/gotypecheck.go
Outdated
| if len(n.Args) > 0 { | ||
| a = n.Args[0] | ||
| } |
There was a problem hiding this comment.
nitpick: len of n.Args can only be 1, right?
There was a problem hiding this comment.
Yes. I should probably validate the syntax of cross() and crossing() while they are still available for check!
|
|
||
| // Validate Gno syntax and type check. | ||
| if err := gno.TypeCheckMemPackageTest(memPkg, m.Store); err != nil { | ||
| return runResult{Error: err.Error()} | ||
| } | ||
|
|
There was a problem hiding this comment.
For my understanding, this is a signal that we’re no longer adding new features to the gno type checker, right?
Do we already have a clear clarification on this plan?
I think what we need to do is review the existing issues related to the gno type checker—either mark them appropriately or close them.
There are also some related PRs, such as #4033. Should we go ahead and merge them? @thehowl @mvertes
There was a problem hiding this comment.
From what jae said, we should go through the PRs and issues relating to the type-checker and label them. I think we can close them for now, then recover them if we want to go ahead and improve our own preprocessor / type checker.
Longer term, I hope that we can migrate away from the GnoVM sooner than we consider removing our reliance on the go type checlker.
No description provided.