This is the mail archive of the guile@cygnus.com mailing list for the guile project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
> > However, if, say, #f > > were the false value and #f\ were the character `f', > > Well, we only need to distinguish #!/ and #! followed by letter. If I'm not mistaken, the canonical magic cookie for a script interpreter is `#! /', with a space, i.e.: #! /usr/bin/perl -w So we'd want to handle that case as well. I'm unhappy with all of the conventions I've seen for handling this matter. It feels as if we've painted ourselves into a corner. To my knowledge, no other scripting language in common use requires any special syntax for #! interpolation, such as the #! ... !# block comment syntax introduced by SCSH. That kind of means that our current convention is a wart. I'd be more comfortable if we could adopt something like: #! as the first two characters in a file represent a comment until end-of-line. ... with some appropriate backward compatibility to handle the #! ... !# syntax that we presently use. Then we could adopt the same argument-processing mechanisms as existing Unix interpreters. I think that if we make the script invocation much more complex than this, it will discourage people from using Guile -- it's a minor inconvenience, but feels very unwieldy. I don't have any strong opinion about how #! should be handled elsewhere in a Guile program. It would be nice if we could invent some meaning that would be consistent throughout the language, but I prefer scripting convenience to consistency for this case. These comments need to be tempered by the fact that I haven't had time to do anything with Guile for about six months, so I am basically providing opinions without doing anything useful about them. Sorry about that. -- T. -- Answers to Frequently Asked Questions: 1. Morgan James Nicholas Pierce. 2. 7 pounds, 12 ounces. 3. http://morgan.rootsweb.com/