Suppress type errors in numeric-input.class.tsx #2433
Conversation
…mittent typechecking errors
|
Size Change: 0 B Total Size: 466 kB ℹ️ View Unchanged
|
npm Snapshot: PublishedGood news!! We've packaged up the latest commit from this PR (6efc9c1) and published it to npm. You Example: pnpm add @khanacademy/perseus@PR2433If you are working in Khan Academy's webapp, you can run: ./dev/tools/bump_perseus_version.js -t PR2433 |
| focus: () => boolean = () => { | ||
| this.inputRef.current?.focus(); | ||
| // TODO(benchristel): type `inputRef` properly. | ||
| // Typechecking fails here intermittently, probably because of nondeterminism |
There was a problem hiding this comment.
I've never had trouble with this, is it possibly a local configuration issue? Are other people seeing this? The fact that TS fails it sometimes seems like a red flag to me. Is there a ticket to follow up on this?
There was a problem hiding this comment.
I don't have any special local configuration. Are you running pnpm tsc --watch from the command line? Or something else?
I seem to recall that @SonicScrewdriver was able to repro this.
There was a problem hiding this comment.
Oh yeah, I was a little confused about this one. It was working perfectly fine on my machine, but I believe other people on the team were seeing some errors. It passed the PR checks fine as well, at the time.
There was a problem hiding this comment.
Still seems fine for me right now. (I checked in both VSCode and Cursor in case Cursor isn't set up properly.)
There was a problem hiding this comment.
The way that we've defined this ref to bridge between two different components feels wrong to me. I can repro the type check failures by using pnpm tsc --watch and then making a change in this file.
I think what we should do is have the ref defined using an interface with the common functions defined in that interface.
|
Superseded by #2452. |
These type errors appeared intermittently when running
tsc --watch. I suspectthere's a cleaner way to deal with these refs. In the short term, suppressing the
errors in code is preferable to tolerating them.
Issue: none
Test plan:
pnpm tsc -w