On Sat, Sep 27, 2025 at 11:31 AM Chris Jones <[email protected]> wrote:
> On 27 Sep 2025, at 5:31 pm, Dave Allured - NOAA Affiliate < > [email protected]> wrote: > > On Sat, Sep 27, 2025 at 10:18 AM Chris Jones <[email protected]> > wrote: > >> On 27 Sep 2025, at 4:51 pm, Dave Allured - NOAA Affiliate via >> macports-dev <[email protected]> wrote: >> >> Github says that a beta version of a macOS 26 runner is available. Can >> this be activated for MacPorts CI? Can build failures on the OS 26 runner >> be temporarily ignored for merging port updates? >> >> https://github.blog/changelog/2025-09-11-actions-macos-26-image-now-in-public-preview/ >> >> I don’t see much advantage to enabling the ci runner only to them ignore >> the build results. If we do this we should do so when we are ready to >> respect the macOS26 build results the same as the other OSes. >> > > Chris, the advantage is rapid debugging turnaround for contributors who do > not have their own macOS 26 system to work on. > > I get that. My point though was I don’t necessarily agree that we should > ignore the results when we enable this OS in the pipelines. > Okay. I was only referring to the requirement of all passing CI's to allow merge. That is only enforced manually by the authorized committers, right? If that is how it works, then I agree with full implementation for this beta 26 runner, with this manual exception.
