Jon Brandvein

IIUC, it sounds like the resolution here from skylib's POV is that we likely will want to put the gazelle dir in another workspace by giving it its own WORKSPACE...

Flag flip tracked by bazelbuild/bazel#8195.

These are actually macros, not rules. The starlark test suites are really just macros that instantiate test cases (rules).

I think the lint intends to prevent you from creating macros without a name argument, because it's usually considered bad form. Certainly that's something we're trying to discourage in google3....

Not sure which CL you mean, I noticed this issue independently of any PR. I'm fine with force-submitting past any lint-only blockers, though if that becomes the habit we should...

This is already tracked upstream by bazelbuild/buildtools#821.

Ideally skylark_library should generate an action that scans its srcs .bzl files for all direct load() targets, and validates that those all appear among the transitive closure of deps. (Indirect...

I didn't realize .bzl files could appear as deps entries. I don't know what other rules do, but I imagine we should validate the .bzl file the same way regardless...

Funny, I had that happen to me when I typed Ctrl+S in the console, but I still had mouse control so I could kill the process by closing the terminal...

Downgrading to P4 to get off my issue tracker since I'm no longer maintaining Python rules. CCing @rickeylev as current label owner. No idea what our stance is on Python...