Hi, Since it's not the first time I'm having a hard time with this issue (and I know others have too), I'm going to ask for advice or for a feature request:
Reading arbitrary Racket-related files with `read' and `read-syntax' becomes complicated when there are #lang and especially #reader lines. As of today, I couldn't find a simple and general way to deal with that. The #lang line can be handled properly with `read-language', but it still eats the first non-whitespace character, thus continuing with `read' will fail, so reading must be restarted from the beginning of the file. The #reader "directive" is more problematic since what follows it can be anything. Does anyone know a way to deal with all that correctly? If not, my feature request would be a more general `read' and `read-syntax' that can read any Racket-based language file and turn it into a s-expr. (It would be even better if comments could be read in something like (struct comment ....) if an optional argument to `read' is provided!) Thank you very much! Laurent
____________________ Racket Users list: http://lists.racket-lang.org/users
