Carl Dehlin
Carl Dehlin
@29th-Day Would you like to add NVCC toolchain to the CI to have it continuously tested, and fix the bug from there?
I think we can close this since there are several workarounds. Implementing a dependency graph solver feels overkill and out of scope for this project. A small suggestion: if your...
Hi @m-mcgowan , thanks for documenting the workaround for this platform. Would you like to update the official documentation as well? I think that is the best way to ensure...
As noted by OP, the feature is implemented by passing the `-d, --duration` flag to the test executable. Problem was just that this is a bit unintuitive. If one wants...
There seem to be a quite high interest in this feature, so I withdraw my statement about wont-fix. I suggest that whoever implements it does it using the reporter API...
I don't think this is a bug. The doctest header is generated and cleaning the build directory should remove it. Or am I missing something?
Ah now I understand, the header target is in the source tree. That is a bit of a wart indeed. Or is it a feature? ;)
I suggest we close since this actually does not affect downstream consumers. If you use doctest as a sub-module and don't tell cmake about the dependency on the generated header,...
@patstew Can you please provide your system information? I can't repro. Thanks! My system: doctest 2.4.8 Ubuntu 22.04 LTS g++ (Ubuntu 11.4.0-1ubuntu1~22.04) Ubuntu clang version 16.0.6
Sorry, just realized my own mistake! Testing against dev branch I also get a crash. The bug manifest itself differently for me though: ```console [doctest] doctest version is "2.4.11" [doctest]...