Keith Smiley

I didn't notice anything in the code about issues that need to be solved before this makes sense to be public, so I was hoping this PR could serve as...

thanks for the context. AFAICT in the open source world, rules_apple doesn't ever use `main_output` https://github.com/bazelbuild/rules_apple/blob/b4a1d1e71ee655ff6d6b51b5845757377febf321/apple/internal/linking_support.bzl#L430-L439 and neither does rules_swift https://github.com/bazelbuild/rules_swift/blob/1aec64c218fc057c2a836e67bd55bc514e0ef8bb/swift/internal/linking.bzl#L612-L624 This must have diverged internally? (even the `upstream` branches...

@comius since it's been a while has there been any movement on your end around this stuff? Looks to me like rules_python is still relying on these private APIs for...

it would be useful to reproduce outside of bazel using clang and relative paths. IIUC right now the logic in bazel is just that it expects all paths (besides system...

> what exactly do you expect to see? if you use `#embed` and also include a header, you should see that the header path is relative, and the embed path...

FWIW I wouldn't be the right person to convince to change anything about the `.d` model, and practically speaking it seems very unlikely to change. But I am also interested...

`additional_compiler_inputs` is intended for this. What did you set embed-dir to?

that actually doesn't seem to help in my test: ``` % clang++-20 -std=c++23 -MM main.cc -no-canonical-prefixes --embed-dir=. -ffile-compilation-dir=. main.o: main.cc /tmp/repro/embed-issue.dat ``` does the repro case from above no longer...

want to rebase?

I would still love to see this land pre 9.0