8 Mar
2025
8 Mar
'25
4:09 p.m.
On Sat Mar 8 14:19:20 2025 +0000, Jinoh Kang wrote:
If CMake could trigger it, it surely isn't an internal behavior. Internal behavior is something that cannot be observed by an application. You should add a test that demonstrates what CMake did that made it run into the bug, not test for the internals directly.
But how to do it in an elegant way? This can be triggered simply in Wine on driver Z: (as / is obviously not writable for a user in Linux) but is not triggered in Wine on C: which is writable in Wine but not on Windows and I expect is the only drive on testbots.
So idea - test on Z: and fallback to C: if Z: is not present?