Duy Nguyen

> I even tempted to make REPL recognizes shell commands; the obstacle is that a command with no arguments can't be distinguished from variable reference. Don't you dare make me...

autoconf has `AC_DEFINE_UNQUOTED`, not sure if the same here (maybe reversed?). Noticed because I had to use this one just the other day.

> Calling main is SRFI-22, which unfortunately wasn't adopted to R7RS Is srfi-22 in r7rs-large agenda? Probably best handled there, I think.

> IMHO the SRFI 22 requirement that main returns an integer is archaic. Returning integers to represent errors is idiomatic in C. This may be related to the POSIX SRFI...

@johnwcowan what about srfi-22 and r7rs? Specifically the `main` function to be called automatically when an r7rs file is run as a script.

Um.. I read srfi-22 again. It requires the shebang line which is not here (at least in Lassik's original report). I think without that line, it's gauche behavior and up...

> Hence it might be worth writing a successor to SRFI 22 that removes the use or numerical exit code from main and uses those executable names in a more...

Yeah. You have some work cut out for you. The situation with r5rs, r6rs and r7rs (if you want to support all) is probably not that different from python 2...

> I'm not sure about whether it extends to code brought into the top level with `include` or not. Most r7rs libraries are split between an ,sci that includes one...

That could help (and I'll look into it). The main problem here though is line-edit clears the screen and redraws whole at every update. When an expression is multiple lines...