Well, thanks to CharTerm, I have made it work. However, in adding some error handling routines for the runtime I've run into a weird bug.
You can see the full code here: https://github.com/jarcane/MicroMini/blob/master/main.rkt The relevant bit, however, is the (crash-handler) function. For some reason that utterly escapes my notice or discovery, the (format "~X" ...) call there does not actually spit out a hexadecimal number at runtime. The weirdest part is: I can't repeat the bug anywhere but when actually running main.rkt. I've even tried hand-entering all the relevant bits into the commandline REPL and calling the crash-handler inside a (with-crashterm), and it doesn't do it there: It properly gives me a hexadecimal number. Other than that it works great so far. Now I need to write at least a basic assembler. On Sun, Aug 10, 2014 at 6:09 PM, Neil Van Dyke <[email protected]> wrote: > If you haven't already, you can try the following to disable any port > buffering within Racket, but that doesn't help you if the OS (Windows and > Posix-ish console stuff it's doing from the console window) isn't giving > Racket raw, unbuffered access to keyboard input through stdin: > > (file-stream-buffer-mode my-port 'none) > > If you were using almost anything other than Windows, you could just use " > http://www.neilvandyke.org/racket-charterm/";. > > Other than that, you or someone could make a portable Racket text-terminal > package that does stuff like uses the console API on Windows and tty on > almost everything else. > > BTW, you mention getting random bytes with variations on "read-bytes". > That simply shouldn't happen with "read-bytes", but, when using variations > on "read-bytes-avail!", only the first N count bytes it says it read will > be valid in the bytestring. > > Neil V. > >
____________________ Racket Users list: http://lists.racket-lang.org/users
