You convince me to your point Axel Wagner. At the same time if we look at
examples in Go Spec, I think their can be improved.
"A0, A1, and []string
A2 and struct{ a, b int }
A3 and int A4, func(int, float64) *[]string, and A5
B0 and C0
D0[int, string] and E0
[]int and []int
struct{ a, b *B5 } and struct{ a, b *B5 }
func(x int, y float64) *[]string, func(int, float64) (result *[]string),
and A5"
I mean, first we need to check that A0, A1 and []string are the same type
and after few examples like D0[int, string] is the same as E0, we have
stated []int and []int are the same type. If you convince yourself that A0
is the same as A1 and both are the same as []string, checking that []int
has the same type as []int is quite trivial. I would prefer that examples
would start from basic cases like []int is []int and []A3 is []int (if this
one is true) and progress to more convoluted like D0[int, string] is E0.
Best regards,
Kamil
czwartek, 4 maja 2023 o 14:12:25 UTC+2 Axel Wagner napisał(a):
> Personally, I'd rather add more examples of "self-evidently equal types".
> In my opinion, all the type aliases in that block confuse matters quite a
> bit.
>
> "[]int and []int are identical" is not actually self-evident at all. It is
> self-evident that any sensible definition of type identity *should* make
> them identical. But it's not self-evident that the given definition *does*.
> Spelling that out in the example, means you are nudged to look at the
> definition and see how their identity follows (by finding "Two slice types
> are identical if they have identical element types").
>
> In fact, whenever you define an equivalence relation, proving that it is
> reflexive is the very first step. And it's not always trivial. For example,
> `==` on `float64` is *not* reflexive. It seems obvious that NaN == NaN
> *should* hold from how it's spelled - but it doesn't.
>
> So, I disagree that the examples should limit themselves to cases where
> it's non-obvious that the two types should be identical.
>
> On Thu, May 4, 2023 at 12:35 PM Kamil Ziemian <kziem...@gmail.com> wrote:
>
>> There is a second such example just below "[]int and []int", but to
>> understand it we need some more type declarations, I listed them below.
>> `type (
>> A0 = []string
>> A1 = A0
>> A2 = struct{ a, b int }
>> A3 = int
>> A4 = func(A3, float64) *A0
>> A5 = func(x int, _ float64) *[]string
>>
>> B0 A0
>> B1 []string
>> B2 struct{ a, b int }
>> B3 struct{ a, c int }
>> B4 func(int, float64) *B0
>> B5 func(x int, y float64) *A1
>>
>> // Unimportant part.
>> )`
>> The line in question is
>> "struct{ a, b *B5 } and struct{ a, b *B5 }"
>> which is true, but again feel out of place. I only start grasping rules
>> of types identity, but I make guess that it should be something like
>> "struct{ a, b *A5 } and struct{ a, b *B5 }"
>>
>> Of course it my just be that I'm just stupid. Feel free to inform me that
>> indeed I have no idea what is going on in the Go Spec.
>>
>> Best regards,
>> Kamil
>> czwartek, 4 maja 2023 o 12:20:35 UTC+2 Kamil Ziemian napisał(a):
>>
>>> Hello,
>>>
>>> In the section "Type identity" of Go Spec we read a list of type
>>> declarations
>>> `type (
>>> A0 = []string
>>> A1 = A0
>>> A2 = struct{ a, b int }
>>> A3 = int
>>> A4 = func(A3, float64) *A0
>>> A5 = func(x int, _ float64) *[]string
>>>
>>> // Part unimportant for my point.
>>> )`
>>> and then we have list of types that are identical. Among them we can
>>> find text
>>> "[]int and []int"
>>> It is obviously true, but feel out of place. I make a humble guess that
>>> authors intended something along the lines
>>> "[]A3 and []int"
>>> Can someone look at this part of Go Spec? I feel that someone make a
>>> mistake, but at the same time humble me saying that there is any mistake in
>>> the Go Spec is something that I shouldn't do.
>>>
>>> Best regards,
>>> Kamil
>>> poniedziałek, 8 listopada 2021 o 10:59:23 UTC+1 Kamil Ziemian napisał(a):
>>>
>>>> Thank you Jan Mercl, now I start to understand this rule.
>>>>
>>>> Best
>>>> Kamil
>>>>
>>>> niedziela, 7 listopada 2021 o 19:34:41 UTC+1 Jan Mercl napisał(a):
>>>>
>>>>> On Sun, Nov 7, 2021 at 7:23 PM Kamil Ziemian <kziem...@gmail.com>
>>>>> wrote:
>>>>>
>>>>> > Can anyone give me explicit example when semicolon is omitted in
>>>>> accordance to the second rule and explanation where it should be? I
>>>>> probably see such situations dozens of times, I just not know that they
>>>>> would needed semicolon in some places.
>>>>>
>>>>> I think this is a simple example:
>>>>> https://play.golang.org/p/ZfKxTos6GjY
>>>>>
>>>>> Click "Run" to see the code is valid, then "Format" to watch one
>>>>> semicolon disappear and then "Run" again to see it's still valid code.
>>>>>
>>>> --
>>
> 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/001d0306-0a43-4680-a03c-3dc87e89dc5an%40googlegroups.com
>>
>> <https://groups.google.com/d/msgid/golang-nuts/001d0306-0a43-4680-a03c-3dc87e89dc5an%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/d31e7546-c489-4b59-a907-a050fc2cce0dn%40googlegroups.com.
