Fabian Meumertzheim

@floriographygoth I think that the docs are missing the shorthand `@foo` for `@foo//:foo`.

@Wyverald If module extensions could depend on repositories defined in other module extensions, I think that this could be used to solve this issue in a very natural way: 1....

This actually works today without any changes to Bazel (although it is affected by https://github.com/bazelbuild/bazel/issues/15916 if the main module defines constraints). As a demo, I created https://github.com/fmeum/local_config_platform, which contains a...

@castler If your actions really care about things being symlinks, you want to use `--experimental_allow_unresolved_symlinks`. Note that this flag doesn't properly work with RBE and runfiles yet - you can...

@tjgq Yes, that's completely correct. I only wanted to draw attention to the fact that even `ctx.actions.declare_symlink` will not work as advertised, although it should be the right choice in...

> If this is really strictly about transition wrappers, which I agree don't work well for this purpose, we could explore custom APIs for that use case, maybe something along...

Maybe the right API for this would be based on [this proposal by @comius](https://docs.google.com/document/d/1uuX1Gz6Kvivp-oOqGja342VrVO534PLF0m_zFtFIBcc/edit#bookmark=id.5ou8nndlopk6): Turn a `Target` into something mutable, mutate it and return it from a rule instead of...

@brentleyjones @coeuvre Would one of you be available to review or suggest a reviewer? It would be really unfortunate if this effort stalled.

https://github.com/bazelbuild/bazel/pull/16008#issuecomment-1224267553 is a first report of a 5.3.0 regression my fix for #16003 introduced - the existing tests didn't cover local execution.

@alexjski While the fate of `BAZEL_TRACK_SOURCE_DIRECTORIES` is undecided, I think that adding a warning about the incorrect incremental behavior of source directories is a necessary step. The warning message is...