Pascal Weisenburger

> the resulting type alias/member either cannot be made abstract, or we should include that information in `Symbol.newType(parent: Symbol, name: String, tpe: Option[TypeRepr])`. Generally, type members define a lower and...

This is another instance of the problem that would be addressed by #20236. So for now I still hope, we can get this fixed.

Thanks for the info! I wasn't aware that this method has a spec. There are two things that are unclear to me: First, the [`computeAsSeenFrom`](https://github.com/scala/scala3/blob/3.4.1/compiler/src/dotty/tools/dotc/core/Denotations.scala#L1082) I adapted does call [`asSeenFrom`](https://github.com/scala/scala3/blob/3.4.1/compiler/src/dotty/tools/dotc/core/TypeOps.scala#L34)...

You are right, in general you should not rely on `underlyingArgument` working in such situations. But I do not think this is an issue in this specific example as `v`...

I rebased on the main branch.

Thanks! 🙂 I assumed this was normal behavior for expanding macro givens, as at some point you probably want a context function `Quotes ?=> T` (to inject the implicit `Quotes`...

> The issue is that if libadwaita application provides its own stylesheets (like e.g. nautilus does), they seem to override any injected debug stylesheets (be it via GTK inspector of...

> I don't think any of those are in [Nautilus' own stylesheet](https://gitlab.gnome.org/GNOME/nautilus/-/blob/master/src/resources/style.css) Some of them are, like the [`border-radius`](https://gitlab.gnome.org/GNOME/nautilus/-/blob/master/src/resources/style.css#L180) of `.nautilus-window listview row` in the example (the selector in the...

Cool, thanks! It would be great if the theme supported those applications :-)