Is there a way to mark testcases in table-driven tests? Like if one of testcases fail it will output in which testcase that happened. Something like Go’s t.Run("testcase name", func(t *testing.T) { ... }).
Let’s say I have this test:
@(test)
bulk_string_test :: proc(t: ^testing.T) {
Testcase :: struct {
input: string,
res: string,
err: bool,
}
testcases := []Testcase{
{ input = "5\r\nhello\r\n", res = "hello", err = false },
{ input = "0\r\n\r\n", res = "", err = false },
{ input = "5\rhello\r\n", res = "", err = true },
{ input = "4\r\nhello\r\n", res = "", err = true },
}
for tt in testcases {
free_all(context.allocator)
r := string_to_stream(tt.input, context.allocator)
res, err := read_bulk_string(r, context.allocator)
if tt.err {
testing.expect(t, err != nil)
continue
}
testing.expect_value(t, err, nil)
testing.expect_value(t, res, tt.res)
w: bytes.Buffer
bytes.buffer_init_allocator(&w, 0, 1024, context.allocator)
testing.expect_value(t, bytes.buffer_to_string(&w), tt.input)
}
}
And it fails with this message.
[INFO ] --- [2025-07-10 21:16:10] Starting test runner with 1 thread. Set with -define:ODIN_TEST_THREADS=n.
[INFO ] --- [2025-07-10 21:16:10] The random seed sent to every test is: 93478856182770. Set with -define:ODIN_TEST_RANDOM_SEED=n.
[INFO ] --- [2025-07-10 21:16:10] Memory tracking is enabled. Tests will log their memory usage if there's an issue.
[INFO ] --- [2025-07-10 21:16:10] < Final Mem/ Total Mem> < Peak Mem> (#Free/Alloc) :: [package.test_name]
[ERROR] --- [2025-07-10 21:16:10] [main.odin:138:bulk_string_test()] expected bytes.buffer_to_string(&w) to be 5
hello
, got
[WARN ] --- [2025-07-10 21:16:10] < 70B/ 1.20KiB> < 1.07KiB> ( 0/ 9) :: main.bulk_string_test
+++ leak 56B @ 0x128420038 [main.odin:143:string_to_stream()]
+++ leak 10B @ 0x128420078 [main.odin:144:string_to_stream()]
+++ leak 4B @ 0x12842008A [main.odin:87:read_bulk_string()]
main [| ] 1 :: [package done] (1 failed)
Finished 1 test in 635µs. The test failed.
- main.bulk_string_test expected bytes.buffer_to_string(&w) to be 5
hello
, got
Currently I need to look at error message and guess for which testcase this output corresponds to. But in some cases it’s not possibles.
I also can see a workaround: include describsion into the msg in every testing.expect. But that won’t be possible for testing.expect_value because it doesn’t have msg argument…