Go - Test break if output invalid utf8 char in TestXXX funcs.

[unifreak]

I'm new to go, and typically learn by running various test cases. I encountered this issue while trying to run codes in test case following https://go.dev/blog/strings.

How to reproduce:

Create a test file with a test func like:

func TestString(t *testing.T) {
	const sample = "\xbd\xb2\x3d\xbc\x20\xe2\x8c\x98"
	t.Log(sample)
}

Then run Go - Test. The test output is empty due to those invalid utf8 characters.

Here is the test output:

> Environment:
>   GOPATH=/home/me/go
> Directory: /home/me/projects/learn/go/ckbk/string
> Command: /usr/local/go/bin/go test -v
> Output:
=== RUN   TestString
> Elapsed: 0.182s
> Result: Success

Expected Output (with my manually created sublime build script or run go test -v in command line, this test case output something like):

=== RUN   TestString
    string_test.go:11: ��=� ⌘
--- PASS: TestString (0.00s)
=== RUN   TestCountRune
    string_test.go:19: 24
    string_test.go:24: 14
    string_test.go:26: 14
--- PASS: TestCountRune (0.00s)
PASS
ok  	learn/go/ckbk/string	0.001s
[Finished in 178ms]

Here is the sublime text trace back when this happen:

command: build
command: show_panel {"panel": "output.golang_build"}
command: insert {"characters": "> Environment:\n>   GOPATH=/home/me/go\n> Directory: /home/me/projects/learn/go/ckbk/string\n> Command: /usr/local/go/bin/go test -v\n> Output:\n"}
Exception in thread Thread-118:
Traceback (most recent call last):
  File "./python3.3/threading.py", line 901, in _bootstrap_inner
  File "./python3.3/threading.py", line 858, in run
  File "/home/hao/.config/sublime-text/Installed Packages/Golang Build.sublime-package/golang_build.py", line 740, in _read_output
    output_queue.put((output_type, chunk.decode('utf-8')))
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xbd in position 23: invalid start byte

command: insert {"characters": "=== RUN   TestString\n"}
command: insert {"characters": "> Elapsed: 0.182s\n> Result: Success"}

This may or may not be any issue, depending on whether there is really a use case outputting invalid utf8 chars when testing.