On Tue, Jun 20, 2017 at 12:46 PM, Will Hawkins <hawki...@borlaugic.com> wrote:
>
> On Jun 20, 2017 3:40 PM, "Ian Lance Taylor" <i...@golang.org> wrote:
>
> On Mon, Jun 19, 2017 at 11:30 PM, Will Hawkins <hawki...@borlaugic.com>
> wrote:
>>
>> On Tuesday, June 20, 2017 at 2:19:25 AM UTC-4, Volker Dobler wrote:
>>>
>>> On Tuesday, 20 June 2017 06:52:58 UTC+2, Will Hawkins wrote:
>>>>
>>>> I know that there is a difference between interface values and dynamic
>>>> types and dynamic values. Is it possible that the documentation about
>>>> the %T
>>>> is misleading? Should it be more specific that it returns the variable's
>>>> "dynamic type."
>>>> [...]
>>>>
>>>> Should the documentation for %T be updated to say that it prints the
>>>> variable's dynamic type?
>>>
>>>
>>> Well, no.
>>> Look at fmt.Printf: It take a format string and a bunch of interface{}
>>> values to
>>> be printed. If %T would print the static type it always would print
>>> interface{}.
>>> To be useful Printf has too look inside it's arguments and inside is a
>>> HelloInt.
>>
>>
>> Mr. Dobler,
>>
>> Thank you for your reply!
>>
>> I agree with your statement that to be useful, the Printf function must do
>> what it does. And, of course I want it to be useful :-)
>>
>> However, I am still concerned about this language from the spec:
>>
>> "The static type (or just type) of a variable is the type given in its
>> declaration, the type provided in the new call or composite literal, or
>> the
>> type of an element of a structured variable."
>>
>> Where it says "just type" means that anywhere the documentation uses the
>> word "type" without qualification, it is referring to the static type. In
>> the case of Printf, then, yes, %T should always print "interface{}"
>> because
>> that is the variable's static type.
>>
>> I 200% (that's more than 100% :-)) agree with you that is not useful.
>> However, based on the language in the spec, it is how I would expect it to
>> operate.
>>
>> I sincerely appreciate your willingness to read through my message and
>> offer
>> your reply. I am already learning very much!
>
> I think you are being misled by trying to use the language in the spec
> to interpret the meaning of %T in the fmt package. Perhaps the fmt
> package docs should be updated, but the actual behavior is not going
> to change. What %T prints is the type of the value passed in, unless
> the value passed in has an interface type, in which case it prints the
> dynamic type of the value.
>
> Ian
>
>
> Thank you so much for your reply. I definitely agree with you and do not
> want the behavior to change at all. I just wanted the language to be
> clarified. As I said in my last email, it seems like any change to the fmt
> documentation would also need to be made to the reflect documentation. The
> word "type" in both of those areas obviously and correctly refers to the
> dynamic type which is in conflict with the definition of the (unqualified)
> term type from the spec. Is there a place that I can advocate for such a
> change?
You can open an issue (https://golang.org/issue) or just send in a
change (see https://golang.org/doc/contribute.html). Thanks.
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+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
