Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
CI has begun failing on MinGW: FAILED ci/tests.py::test_existence[gvmap.sh] - AssertionError: gvmap.sh has been resurrected in the cmake build on None. Please remove skip. assert WindowsPath('C:/Graphviz/bin/gvmap.sh') is None + where WindowsPath('C:/Graphviz/bin/gvmap.sh') = which('gvmap.sh') FAILED tests/test_tools.py::test_tools[gvmap.sh] - OSError: [WinError 193] %1 is not a valid Win32 application Comparing the CI logs from commit 71aed63 against recent failing MRs, it can be seen that Python in MinGW has moved from 3.11.10 to 3.12.7. Quoting some text from commit 97b0c0b: > This test began failing due to gvmap.sh unexpectedly existing. As diagnosed > by Magnus: > > > I diffed a full log file from a successful job with one from a failing one > > and the only tangible difference I could find was an upgrade of Python > > from 3.11.5 to 3.12.0. Indeed, this is what What’s New In Python 3.12¹ > > says: > > > > > shutil > > > … > > > shutil.which() now consults the `PATHEXT` environment variable to find > > > matches within `PATH` on Windows even when the given `cmd` includes a > > > directory component. (Contributed by Charles Machalow in gh-103179.) > > > > > > shutil.which() will call `NeedCurrentDirectoryForExePathW` when querying > > > for executables on Windows to determine if the current working directory > > > should be prepended to the search path. (Contributed by Charles Machalow > > > in gh-103179.) > > > > > > shutil.which() will return a path matching the `cmd` with a component > > > from `PATHEXT` prior to a direct match elsewhere in the search path on > > > Windows. (Contributed by Charles Machalow in gh-103179.) > > > > I don't really understand what all this means, but it sounds quite > > suspicious to me anyway. So it looks like the previous Python change that made it into the Windows ecosystem has now propagated to MinGW.
- Loading branch information