Exemption for for...of unused variable.

[dimaqq]

In similar vein to for...in #2342
Should there be similar exemption for [async] for...of?

What rule do you want to change?
no-unused-vars

Does this change cause the rule to produce more or fewer warnings?
fewer warnings

How will the change be implemented? (New option, new default behavior, etc.)?
I don't know

Please provide some example code that this change will affect:

async function forever(x) {
  for await (const unused of x) {
    void unused;
  }
}

forever(monitor_network());

What does the rule currently do for this code?
Line 42: 'unused' is defined but never used no-unused-vars

What will the rule do after it's changed?
Nothing

Are you willing to submit a pull request to implement this change?
This seems tricky...

[mqklin]

This is broken for me after eslint update to v6 as well:

for (const x of y) { // x is defined but never used (no-unused-vars)
  console.log(x);
}

[kaicataldo]

This looks like a duplicate of #12117. Can you see if the solutions there fix your issue?

[kaicataldo]

Note that this issue doesn't show in the ESLint Demo .

Are you using babel-eslint?

[mqklin]

Yes, I use babel-eslint. Yes, updating babel-eslint fixed the issue, thanks a lot!

[g-plane]

Closing this issue as it looks like the question has been answered. Please feel free to visit us in the ESLint Gitter if you have any other issues!

[platinumazure]

@g-plane I could be missing something, but... I'm not sure this should be closed. The original post was a rule enhancement request for "for await of", but then someone else came and asked about "for of" and we noted the babel-eslint/eslint-scope issue.

Should we still evaluate the original rule request? If not, why not?

[g-plane]

Oh, it seems reasonable. Sorry.

[kaicataldo]

I'm not able to reproduce with our demo.

@dimaqq I see you left a reaction on my comment above - did this solve your problem (hoping it did!)?

[dimaqq]

I can only test it on Monday, sorry.
I think that with void trick it ought to work; and without it I’m not sure...

[mysticatea]

I guess you can use varsIgnorePattern option to ignore unused variables by naming convention.

For example, Online Demo.

[dimaqq]

There's something wrong with the online demo, it says Lint-free even for code like this:

function forever(x, y) {
  const z = 42;
}

https://eslint.org/demo#eyJ0ZXh0IjoiZnVuY3Rpb24gZm9yZXZlcih4LCB5KSB7XG4gIGNvbnN0IHogPSA0Mjtcbn0iLCJvcHRpb25zIjp7InBhcnNlck9wdGlvbnMiOnsiZWNtYVZlcnNpb24iOjExLCJzb3VyY2VUeXBlIjoibW9kdWxlIiwiZWNtYUZlYXR1cmVzIjp7fX0sInJ1bGVzIjp7fSwiZW52Ijp7fX19

[kaicataldo]

None of the rules are on in the link you provided. When I turn on no-unused-vars, it’s displaying an error as expected.

[dimaqq]

I'm so sorry for the confusion.

When I read #2342 I was left with the impression that the following is allowed:

for (unused in something) {}

But of course that's only because unused could be anything here, e.g. a global, and other rules kick in now: no-undef, prefer-const.

It didn't think that the following is semantically different wrt. the unused variable rule and treated both cases the same:

for (const unused in something) {}

Weirdly there's test case for the above, as if that's valid code:

eslint/tests/lib/rules/no-unused-vars.js

Line 159 in 76fb571

{ code: "(function(obj) { for ( const name in obj ) { return true } })({})", parserOptions: { ecmaVersion: 6 } },

Yet, if I "enable all rules" in the eslint demo, that very example triggers no-unused-vars 😕

In any case void unused; works and keeps eslint silent.

Also, as far I can tell for await (const x of y) is treated exactly same as for (const x of y).

I guess I'm OK then :)

[kaicataldo]

Thanks for following up! I'm going to go ahead and close this issue then.