home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!snorkelwacker.mit.edu!ai-lab!zurich.ai.mit.edu!jinx
- From: jinx@zurich.ai.mit.edu (Guillermo J. Rozas)
- Newsgroups: comp.lang.scheme
- Subject: Re: wots going on here!?
- Message-ID: <JINX.92Sep2121245@rolex.ai.mit.edu>
- Date: 2 Sep 92 16:12:45 GMT
- References: <17timvINN7b8@agate.berkeley.edu> <17tn1uINNkij@early-bird.think.com>
- <JINX.92Aug31170432@chamarti.ai.mit.edu>
- <17ur1oINNem6@agate.berkeley.edu>
- <JINX.92Sep1033012@chamarti.ai.mit.edu>
- <SOMMERFELD.92Sep1131757@gourmet.apollo.hp.com>
- <t6jnlga.thinman
- Sender: news@ai.mit.edu
- Reply-To: jinx@zurich.ai.mit.edu
- Organization: M.I.T. Artificial Intelligence Lab.
- Lines: 21
- In-reply-to: thinman@netcom.com's message of 1 Sep 92 23:12:54 GMT
-
- In article <t6jnlga.thinman@netcom.com> thinman@netcom.com (Technically Sweet) writes:
-
- | sommerfeld@apollo.hp.com (Bill Sommerfeld) writes:
- | >"#[unspecified IF alternative at line xxx character yyy of source file zzz.scm]"
- |
- | >or some other such reference back to the source code which would point
- | >the beginner right at the bug.
- |
- | >These could be generated *once* when reading in the source.
- |
- | The old BSD Pascal interpreter/compiler used this convention:
- |
- | line xx of file yy: unknown whoziwhatsis:
- | a := b + c; d = 3;
- | ----------------------^
- |
- | This is still the best error reporting I've seen.
- | I used it on a recent interpreter project.
-
- That is fine (and desirable) for syntactic errors. Much harder (when
- at all possible) for runtime errors.
-