Re: Boost Build testing framework conundrum
Edward Diener via Boost-build <[email protected]> Tue, 28 Apr 2020 13:56:45 -0400
| Newsgroups | gmane.comp.lib.boost.build |
|---|---|
| Message-ID | <[email protected]> |
On 4/28/2020 7:03 AM, Steven Watanabe via Boost-build wrote: > AMDG > > On 4/22/20 10:36 AM, Edward Diener via Boost-build wrote: >> Does the Boost Build testing framework, when running the >> 'toolset_clang_linux.py" test, actually invoke the clang compiler when >> running the test ? Because when I run the test on my local machine in >> Windows 10 the clang compiler is not even in my PATH at the time the >> test is being run, yet the test "succeeds" on the current 'develop' >> branch whereas for my PR it is failing. So I do not understand how the >> test can verify that the clang-linux.jam is correct. > > The test uses a set of hard-coded command lines. Please see my other post which specifies a bug in the Boost Build testing framework when running the 'toolset_clang_linux.py' test. Since you are the author of that test, if you could get a chance to look at it and correct the bug it would be greatly appreciated. > >> Can someone please >> enlighten me on how this is supposed to work ? I do run the tests with >> the --preserve option but when the test succeeds the entire directory >> structure where the files would be is in a temporary directory and all >> the files in that temporary directory are removed by the end of the test. >> > > In Christ, > Steven Watanabe _______________________________________________ Unsubscribe & other changes: https://lists.boost.org/mailman/listinfo.cgi/boost-build