Quirin F. Schroll

It might be a late addition, but @tgehr convinced me in several private and public conversations that it’s much more orthogonal and valuable to add “attribute variables”, i.e. a something...

Shouldn’t the `@called` attribute be a parameter attribute (along the lines of `scope`)? A function attribute feels wrong. It’s not really part of the function type.

The respective [Unicode Block](https://en.wikipedia.org/wiki/Mathematical_Alphanumeric_Symbols) does not include different cursive scripts. Libertinus Math could only support them via some style option, but probably won’t. One solution would be to use a...

It’s not part of a Unicode block. It’s a way to add them, though.

@mdparker, at least, add me. As far as I can tell, this change is neutral for most people and mildly* valuable for those with privacy concerns. As stupid as it...

> Why is there a new branch for function pointers, is the final `else` case obsolete now? Because `toDelegate` works with any callable, the `else` branch cannot be `@safe` in...

> > * $(LI Does not work with `@safe` functions.) > > This line is no longer accurate now That should have been a review comment.

Sorry, I was absent for a while. I looked into the failed checks and didn’t understand what’s wrong. Should I force-push?

@dukc, thank you for reading. > However, I'm less convinced about the `@nodbi` attribute. It solves a problem that can be mostly solved by simply assigning an `enum` parameter to...

What does this PR do exactly? If it makes casting a C++ class impossible, there is no way to fix code. There must be a way to `static cast` C++...