On Tue Jun 14 04:03:29 2022 +0000, **** wrote:
Zebediah Figura replied on the mailing list: ``` On 6/13/22 03:27, Rémi Bernon (@rbernon) wrote:
Rémi Bernon (@rbernon) commented about dlls/user32/tests/input.c:
ret = GetRawInputData((HRAWINPUT)lparam, RID_INPUT, &raw,
&raw_size, sizeof(RAWINPUTHEADER));
ok(ret > 0 && ret != (UINT)-1, "GetRawInputData failed\n"); ok(raw.header.dwType == RIM_TYPEMOUSE, "Unexpected rawinput
type: %lu\n", raw.header.dwType);
+ ok(raw.header.dwSize == raw_size, "Expected size %u, got %lu\n", raw_size, raw.header.dwSize); + todo_wine_if (wparam) + ok(raw.header.wParam == wparam, "Expected wparam %Iu, got %Iu\n", wparam, raw.header.wParam); We don't usually indent `todo_wine_(if)`, do we?
Some people do. Semantically "todo_wine" is similar to "if (x)" [and "todo_wine_if" is similar to "if"] so I indent it accordingly. This also has the advantage that the line will be touched again when the todo_wine is removed, which can make it easy to spot what commit fixed a test in `git blame` (where applicable). ``` Sure, it's a matter of taste, but it didn't seem to me that we _usually_ indent it. So that for instance, removing the todo doesn't need to touch the test line itself.
-- https://gitlab.winehq.org/wine/wine/-/merge_requests/233#note_2013