doc/go1.15: Vet notes typo

[MaerF0x0]

link: https://tip.golang.org/doc/go1.15#vet

What did you expect to see?

Allow string conversion from byte slice

to only permit string(x) for integer x when the type of x is rune or byte slice.

What did you see instead?

We are considering prohibiting the conversion in a future release of Go. That is, the language would change to only permit string(x) for integer x when the type of x is rune or byte. Such a language change would not be backward compatible. We are using this vet check as a first trial step toward changing the language.

I didnt know where to submit a CL

[dmitshur]

Thanks for reporting.

My understanding is that the current text is correct and intentional. It refers to conversions like string(x) where x is not a slice type, but rather a rune or byte. Conversions of []byte to string is not the topic being discussed; that's out of scope of #3939.

I'll let @ianlancetaylor who authored CL 234517 confirm. If it turns out more people find the current wording confusing, we can try to improve it.

[ianlancetaylor]

Yes, the existing text is correct. Thanks for asking about it.

[MaerF0x0]

I see, yes. Prompted by your notes as I read it closer I came to understand the text block is only talking about for integer x. My incorrect reading of text had me concerned I'd lose the ability to convert []byte when I want to print as strings, something like:

data, _ := json.Marshal(map[string]interface{}{"JSON": "Yes"})
// data is []byte
fmt.Println("json is", string(data))