Adrian Taylor

Thanks for the report. If you'd like me to look at this any time soon, please raise a pull request with a failing test in `integration_tests.rs`. I'm not actually sure...

Per the discussion on #1394 I think there's a bindgen issue to be addressed here first, so closing this autocxx issue.

Thanks for getting in touch. I can see that this would indeed be pretty painful. autocxx tries really hard to stick to the basic interop model of cxx, and exception...

So, autocxx fundamentally works a bit like this: 1. Run bindgen to generate bindgen-like bindings 2. Post-process them to generate input to cxx 3. Feed that into cxx So, I...

> * Name canonicalization -- I imagine a user specifying `throws!("Klass::method")`, but the parts of the engine where function return type conversion happen seem to have a post-bindgen view of...

Please file a test case [PR following these instructions](https://google.github.io/autocxx/contributing.html). Unfortunately all the stuff around names in autocxx is very fragile so this might not be an easy fix.

I'd accept a PR to fix this. To be honest just having a PR which I can use to reproduce the problem in a test case might be enough for...

Thanks - I'm glad you found a workaround. As I mentioned, the naming stuff is annoyingly fragile, and unfortunately I think providing custom names for stuff falls into the category...

I've now dug into #520 enough to refresh my memory of the naming stuff, and what it would take to fix this. I'm not likely to tackle this myself, but...

The PR to reject these names didn't turn out simple. More significant progress on #520 is required here.