fmt.Printf("%f\\n",x) 将 Float64 舍入为 Float32，但 fmt.Println(x) 不会

Question

I was working on some tests with dictionaries, to do this I print out the Float64 values I want from a database in a format to copy and paste them into my test struct array, but when my tests failed I noticed that the values differ, but only by 0.0000000000002

Then, to check the value I wrote the following in a loop:

fmt.Printf("%f\n",value)
fmt.Println(value)

And I got the following values back:

702.200000
702.1999999999998
5683.090000
5683.089999999998
975.300000
975.3

I checked the docs and saw nothing that indicates that there's special notation for Float64 or that %f will replace Float64 for a Float32, however, I don't get the problem when I use %v or %g and the docs specify that they'll use %f when appropriate.

I don't get the problem when I specify a precision of 12 using %.12f either, but no default precision is specified in the docs.

Why does this happen?

EDIT: This is the same issue as the duplicate, but I believe Adrien's explanation about it is more detailed.

Answer 1

From the docs :

The default precision for %e, %f and %#g is 6; for %g it is the smallest number of digits necessary to identify the value uniquely.