Understood! I appreciate all your help.
@Axel Wagner @Jason Phillips @Ian Lance Taylor Thanks again! Xudong 在2024年7月17日星期三 UTC+8 14:02:07<Axel Wagner> 写道： > Hi, > >> However, I would like to point out that selector rule 3 states: >> >> As an exception, if the type of x is a defined pointer type and (*x).f is >> a valid selector expression denoting a field (but not a method), then x.f >> is shorthand for (**x).f.* >> >> The emphasis here is on "defined pointer type." In the code, *s is a >> composite pointer literal type, not a defined pointer type. Therefore, I >> believe this rule does not apply in this case. >> > As I mentioned on slack, the fact that this is not a defined pointer type > means that is not the right rule to look at. Instead, you need to look at > the first rule in the list. > >> Furthermore, in selector rule 1, it only states that "x.f denotes the >> field or method." The term "denotes" is somewhat ambiguous and does not >> imply implicit pointer indirection. >> > I don't think this particular level of language-lawyering is useful. The > spec is not a legal document or anything like that and should be read for > intent. > Writing `f().i` is allowed and well-defined, so the goal should be to > justify *that* it is correct, not to come up with readings making it > incorrect. A gracious reading of the spec here seems pretty clear. > > >> 在2024年7月17日星期三 UTC+8 12:51:48<Ian Lance Taylor> 写道： >> >>> On Tue, Jul 16, 2024 at 9:29 PM 王旭东 <wxd...@gmail.com> wrote: >>> > >>> > Hi, I have a question about "addressable." >>> > >>> > The following code illustrates my question (see the comments): >>> > https://go.dev/play/p/xpiPXuEqh0O?v=gotip >>> > >>> > >>> > >>> > I also posted question in the Gopher Slack channel, and @Axel Wagner >>> provided a detailed explanation, suggesting that this might be a gap in the >>> Go specification. >>> > @Jason Phillips recommended that I ask this question in the GitHub >>> issues and the golang-nuts mailing list. >>> > >>> > I really appreciate everyone’s help mentioned, but I still don’t have >>> a clear conclusion. >>> > >>> > To simplify: my question is why the result of myfunReturnPointer() is >>> addressable if we strictly follow the specification. >>> >>> >>> In https://go.dev/ref/spec#Selectors the spec explains that if the >>> type of x is a pointer, then the selector expression x.f is shorthand >>> for (*x).f. That is the case in your playground example: the >>> expression myfunReturnPointer().i is short for >>> (*myfunReturnPointer()).i. The pointer indirection >>> *myfunReturnPointer() is addressable according to >>> https://go.dev/ref/spec#Address_operators. So this is a field >>> selector of an addressable struct operand. >>> >>> Ian >>> >> -- >> > You received this message because you are subscribed to the Google Groups >> "golang-nuts" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to golang-nuts...@googlegroups.com. >> > To view this discussion on the web visit >> https://groups.google.com/d/msgid/golang-nuts/0fb5aa3e-4568-4520-bd42-ff01c1a2bd54n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/golang-nuts/0fb5aa3e-4568-4520-bd42-ff01c1a2bd54n%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/c4c460a9-cdda-46d1-b5fa-a9dac510a8b7n%40googlegroups.com.
