About the confusion caused by type conversion in golang

[Barry Li]

Some doubts about type conversion in go.

In the following code, why can it be executed normally after passing the variable?

The code is an error during the compilation phase.

// The following two lines of code can be executed normally

    x := uint64(256)

    fmt.Println(byte(uint64(x))) // normal execution

// The following code cannot be executed normally

    fmt.Println(byte(uint64(256))) // constant 256 overflows byte

Can anyone help me answer it?

[Ian Lance Taylor]

In Go, arithmetic on constant values follows different rules than
arithmetic on variable values. For example, as you've seen, an
out-of-range conversion is a compilation error for constants but a run
time error for variables. The same is true of division by zero. See
https://golang.org/ref/spec#Constant_expressions.

Ian

[Arunkumar Gudelli]

256 considered as `constant`. (untyped)

So when we try to convert to byte (which has a range from 0 to 255), it will throw the overflow error.

Where `x := uint64(256)` we are giving a type to constant uint64. 

y := byte(256) // Same error

so byte(uint64) will be zero without any errors.

regards,

Arun Gudelli

https://golangtutorial.dev/

--
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/401197ca-1cdc-45e8-9135-bafe0a300e21n%40googlegroups.com.

[Rob Pike]

This topic is well explained in the blog post at blog.golang.org/constants, if I do say so myself.

-rob

To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/CAOOGzbXe84WeaN2rE7POow5gs_T5JikJfH9Vh8_EruEK_C3bKw%40mail.gmail.com.