Nobody has reached out to me about teaching at SCA Expo in Boston, but one of the classes on offer has exactly the same description as a class I developed ages ago under a new title. (I wrote the class with the express intent that lots of other people would be able to pick up the material and teach it, and many people have so don't read this as me taking issue with that. I kind of hope it is just what it looks like.)
(all of the competing programs I've looked at on this front do worse, but I don't think that's a good reason for me to not try to do better)
It's been a while since I used goto in C++ but I don't feel the slightest bit bad about doing that today. The code I'm working on now is going to get refactored pretty much immediately once it's done so I probably won't need that in whatever this turns into when it ships and even if I did, well, the project I'm working on has been open source for nearly 2 decades without a single outside code contribution so I'll do what I want.
The too much work option is to just build something custom myself, which I could, but that's not a yak I need to shave.
The option that upstream is working on I might want to switch to because it's better, but I might not because it brings in too many dependencies for something most of the people using my software will only very rarely need to interact with (I'm probably the heaviest user of the feature).
Very little coding today. For the next thing I want to work on I identified 5 ways to do the first step. 2 of them were as near as I can tell the exact same thing accessed very slightly differently for compatibility reasons (hate it, not using it). One was what the previous 2 should have been implemented in terms of but wasn't, presumably for dependency management reasons (less hated, what I'm going with for now). One is not really an option but upstream is working on it. Last is too much work.
Author of Typica software for coffee roasters.