All right, I think I'll stop here. Parsing the -U0 ranges in raw C++ is gonna be too much wasted effort. I'll finish this another time, likely after switching to Go.
For historical reasons, mainly. If I were starting from scratch today, I wouldn't pick it. I'm in the process of making a change to golang, but I want to use the existing project to develop the next-gen version. There's only a few more things I need to add to make it viable for working on a large-scale project.
I started the project a year ago. I wanted to have the power and flexibility to achieve anything, and use OpenGL for graphics so there'd be no restrictions as to what can be put on screen. Plus I've been using C++ for many years before for making indie games.
Back then, I was even more motivated to work on it because I felt C++ itself badly needed a more efficient way or writing it. I really dislike the manual duplication it forces you to create and maintain as you change things (e.g. cpp and header files). But even with a more script-like high level language like Go, I feel there is a lot of potential to innovate past a basic text editor. I'm gonna find out.
But I was able to get a rough prototype running.
Screenshot: https://dl.dropbox.com/u/8554242/dmitri/projects/Conception/...