Earlier open issue and discussion about adding a vet check #16520

Earlier open issue and discussion about adding a vet check #16520

Note that #16520 is about improving the vet check. This proposal is to add the already existing check to the default set run by go test.

Can someone characterize the false positives for loopclosure?
The usual rule for turning on a vet check during go test is that it
have basically no false positives, or the false positives are only
in code that you couldn't defend not rewriting anyway.

The docs in loopclosure.go say:

This analyzer checks for references to loop variables from within a
function literal inside the loop body. It checks only instances where
the function literal is called in a defer or go statement that is the
last statement in the loop body, as otherwise we would need whole
program analysis.

For example:

        for i, v := range s {
                go func() {
                        println(i, v) // not what you might expect
                }()
        }

See: https://golang.org/doc/go_faq.html#closures_and_goroutines

In this case it seems like the reason it's not on by default is it doesn't catch very much. It does catch this one example. Maybe that's enough?

Damien, you said that we've been running it inside Google and also claimed that it catches a common mistake, but it looks like it will flag very few of the actual mistakes I've seen, because of the restriction about looking only at the final statement of the loop. Do you know if this has caught real mistakes inside Google, and if so how many?

Timed out in state WaitingForInfo. Closing.

(I am just a bot, though. Please speak up if this is a mistake or you have the requested information.)