On Thu, Mar 29, 2018 at 10:13 AM, Nilsocket <nilsoc...@gmail.com> wrote:
>
>> > https://research.swtch.com/godata
>
> I did read it, but what I was asking is irrelevent to it.
I guess I don't see why. It seems to exactly answer your question.
> func showInt(x int) {
> res := (*[unsafe.Sizeof(x)]byte)(unsafe.Pointer(&x))
> for i := range res {
> fmt.Printf("%.2x ", res[i])
> }
> fmt.Println()
> }
> // showInt(12345)
> // 39 30 00 00 00 00 00 00
>
> An integer in it's binary form is stored like above, little endian.
> What I mean is how, strings are represented, while converting a string to
> []byte, gives me utf-8 values,
> but directly ascessing the underlying memory of string, without converting
> it to []byte shows different things.
Because, as described at research.swtch.com/godata, a string is stored
as a pointer and a length, and when you use unsafe to look at the
contents of a string value you are looking at the pointer and the
length. If you want to look at the actual contents of the string--the
bytes that correspond to the bytes stored in a []byte--you need to
indirect through the string pointer. See also reflect.StringHeader.
>> > Because in Go the size of an array must be a constant.
>> >
>> > Ian
>
>
>
> An array size must be constant, but my real question is?
> How come, a uintptr type value returned by unsafe.Sizeof(x), is considered
> as const?
> and doesn't accept, if uintptr type value is returned by someother function.
> I mean, how was it able to differentiate between someother function and
> unsafe.Sizeof(x).
Because unsafe.Sizeof is a special case, as documented at
https://golang.org/ref/spec#Package_unsafe .
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.
