Float variable is casted to float64

[Archos]

Why when is assigned a float number to a variable (a := 1.5), it is
casted auto. to float64 instead of to be casted to float32 (since it's
a small value)?


===
package main

import (
"fmt"
)

func main() {
a := 1.5
fmt.Printf("%T : %v\n", a,a)
}
===

[Archos]

float64 : 1.5

[Jan Mercl]

It's all in the specs.

http://golang.org/doc/go_spec.html#Variable_declarations

If the type is present, each variable is given that type. Otherwise, the types are deduced from the assignment of the expression list.

If the type is absent and the corresponding expression evaluates to an untyped constant, the type of the declared variable is bool, int, float64, or string respectively, depending on whether the value is a boolean, integer, floating-point, or string constant:

var b = true    // t has type bool
var i = 0       // i has type int
var f = 3.0     // f has type float64 

var s = "OMDB" // s has type string

http://golang.org/doc/go_spec.html#Short_variable_declarations

A short variable declaration uses the syntax:

ShortVarDecl = IdentifierList ":=" ExpressionList .

It is a shorthand for a regular variable declaration with initializer expressions but no types:

"var" IdentifierList = ExpressionList .

 

[Peter Bourgon]

Furthermore, using float64 as the default was a decision made ~a year
ago, for reasons largely to do with the math pkg:

http://groups.google.com/group/golang-nuts/browse_thread/thread/25d8068784971c15/cd41a810152b3d23
http://groups.google.com/group/golang-nuts/browse_thread/thread/d7545b8325c43e15/a689185d647f54e5

[⚛]

On Oct 17, 12:26 pm, Archos <raul....@sent.com> wrote:
> Why when is assigned a float number to a variable (a := 1.5), it is
> casted auto. to float64 instead of to be casted to float32 (since it's
> a small value)?

Decimal numbers such as 0.1 cannot be expressed accurately in binary
form. So from precision viewpoint, float64 is slightly better than
float32.

[j...@webmaster.ms]

Also complex128, it seems missed from the manual.

  var aa = 1+2i

  fmt.Printf("%T : %v\n", aa,aa)

[Jan Mercl]

On Monday, October 17, 2011 2:13:24 PM UTC+2, j...@webmaster.ms wrote:

Also complex128, it seems missed from the manual.

  var aa = 1+2i

  fmt.Printf("%T : %v\n", aa,aa)

Thanks, issue filled: http://code.google.com/p/go/issues/detail?id=2377 

[Archos]

Then, it is better to use directly float64 instead of float32, isn't
it? Iin our programs that handle float numbers.

[Kyle Lemons]

Then, it is better to use directly float64 instead of float32, isn't
it? Iin our programs that handle float numbers.

Float32 becomes insufficient much faster than you might think.  You don't see "float" very often in C, you almost always see "double."

~K