eht16 left a comment (geany/geany#4472) > > Note that this is actually a WIP for now, because I'm not sure whether the > > tags match works, nor whether the if guards do. If you happen to know > > that'd be nice, otherwise I probably will try and create a dummy repo to > > test the tags thingy. > > I did create a test repo, and it seems top work nicely: > https://github.com/b4n/test-workflows/actions I realized I possibly needed to > be able to sign on manual workflow runs, so I added an option (in current > [4db9d47](https://github.com/geany/geany/commit/4db9d47b754fb863da5747006289c820c48ae288) > but that commit ID is likely to change on future rebasings)
:+1: > > @giuspen could you by and chance give Meson on MSYS2 a try and see if it's > > release material or what should be fixed/added? > > […] > > However, the NSIS fails, so I can assume the current Meson result is not > > sufficiently identical to the Autotools one to be a drop-in replacement for > > the moment. What should be fixed I don't yet know :) > > So I also played a bit with the Windows MSYS2 support, and it should be > pretty good now (and the build succeeds!). @giuspen and/or @eht16 could you > give the resulting installers a look to see if it's actually working, not > just _looking_ OK from afar? I cannot test it unfortunately due to no supported operating system :smile:, maybe @techee? > * no generated icon cache for the Tango icons (but that's the same on all > platforms) > > I can fairly easily fix the cache generation by manually running > `gtk-update-icon-cache` on the Tango directory `gnome.post_install()` doesn't > look at, but I'm not even sure how badly this is needed -- and that would > alter behavior on all I think to remember it was necessary but this should be verified again to get sure. It could also be that there was another problem which I didn't see or there is a better fix than this. > Anyway, all the Meson changes are probably gonna end up in a separate PR once > I merge the gist of that one (running the MSYS2 workflow automatically), so > we can discuss some details later as well. > > @giuspen @eht16 once you give this a look, The code changes look fine to me, as I said above, I cannot test it :(. > I'll probably remove the failing cross build runs, unless anybody can find a > solution for it. Opinions? To be honest, I didn't even look into the failing cross build runs when I see your great attempt here and I hope we can simply drop the cross build workflow. -- Reply to this email directly or view it on GitHub: https://github.com/geany/geany/pull/4472#issuecomment-3605901957 You are receiving this because you are subscribed to this thread. Message ID: <geany/geany/pull/4472/[email protected]>
