Saem Ghani

You're figuring it out, keep going, @canelhasmateus

@canelhasmateus should this go back into drafts while you keep plugging away?

Can we mark push and pop as in the danger zone of being axed or heavily changed? I think they're really problematic implementation wise and the syntax doesn't get enough...

oh goodie, here is a fun bit: If the user sets warnings to be raised as errors then I think those warnings need to be within the AST in a...

@haxscramper took a bit but to a first crack at it. Wrote it quickly on my phone during lunch so hopefully it's not too bad.

Thank you very much, @zerbina. That was really helpful. 🎉 I think your analysis is correct, from what I can tell only the `rintIce`/`lexDiagInternalError` are "fatal", all other errors progress...

Sounds good thanks. I'll try that approach first.

@zerbina I pushed the latest, if you're curious. It turns out that only truly fatal stuff (`internalError`) should raise, because the parser/lexer rely on not raising due to `check`. I...

_for the curious: my thoughts as I'm working along_ ### This PR and the Near Term Right now I'm putting all the bridging code for module specific diagnostics to `Report`...

For `nimconf` - I created a type for a callback in `nimconf` - I'll have the two exported procs in `nimconf` that need reporting accept callback as an argument -...