vtjnash wrote: > I'd really rather avoid installing doxygen from a contributor's fork.
I'd be happy to not host also, though I don't foresee any actual issues. An admin can either transfer my fork to the llvm project, or just copy the binary download to somewhere controlled by LLVM infrastructure (e.g. the download server). The content is being served by github currently, not actually me, and at most a malicious binary–if someone hacked my account–would do is run inside the LLVM CI, which is no worse than any random PR already does. The files are currently still available as GHA artifacts, so you also can see they were built by an entirely reproducible process. For further traceability, I've added a record here of the sha256 checksums of the files: [macOS.Latest.Release.Apple.Silicon.build.artifacts.zip](https://github.com/vtjnash/doxygen/releases/download/Dev_1_17_0/macOS.Latest.Release.Apple.Silicon.build.artifacts.zip) sha256:a90172008a5a8774f3cbf4d1f5b5e6b42073feb5b55fbd819f47e9d97aa4cb3e [macOS.Latest.Release.Intel.build.artifacts.zip](https://github.com/vtjnash/doxygen/releases/download/Dev_1_17_0/macOS.Latest.Release.Intel.build.artifacts.zip) sha256:9b5cde661207f797b3257672dab8328e01f3ed472439a93a98857e329e8b7bdc [Ubuntu.Latest.GCC.Release.Arm.build.artifacts.zip](https://github.com/vtjnash/doxygen/releases/download/Dev_1_17_0/Ubuntu.Latest.GCC.Release.Arm.build.artifacts.zip) sha256:d8140fa304ee2dca7f99223a4cd6f5e2e3f0ae7a1986333f896995fa0b8e278b [Ubuntu.Latest.GCC.Release.Intel.build.artifacts.zip](https://github.com/vtjnash/doxygen/releases/download/Dev_1_17_0/Ubuntu.Latest.GCC.Release.Intel.build.artifacts.zip) sha256:76eac676725abbcbf09bc2592808f915394ed291f889583d414493ac3fd862c7 [Windows.Latest.MSVC.Release.build.artifacts.zip](https://github.com/vtjnash/doxygen/releases/download/Dev_1_17_0/Windows.Latest.MSVC.Release.build.artifacts.zip) sha256:37cb6259964008695d2a223a11089193b0e12864afa75d30cb1017a42a054ff8 > Long term we should also really be pushing to switch to clang-doc over > doxygen. Seems fine as a plan. Someday that transition will perhaps happen, and you can delete this. I don't see the relevance to whether we first improve the experience for users of the existing versions of LLVM? https://github.com/llvm/llvm-project/pull/191501 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
