home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.scheme
- Path: sparky!uunet!destroyer!gatech!bloom-beacon!INTERNET!dont-send-mail-to-path-lines
- From: cph@martigny.ai.mit.EDU (Chris Hanson)
- Subject: wots going on here!?
- Message-ID: <9208311208.aa23222@mc.lcs.mit.edu>
- Sender: daemon@athena.mit.edu (Mr Background)
- Organization: The Internet
- References: <1992Aug30.182548.13079@pasteur.Berkeley.EDU>
- Date: Mon, 31 Aug 1992 16:13:02 GMT
- Lines: 19
-
- Date: Sun, 30 Aug 1992 18:25:48 GMT
- From: Matt Wright <matt@volga.berkeley.edu>
-
- But I agree that's it's a tradeoff. I *have* been in the situation where I
- generated an error and wished I could get back to that context, but I got
- thrown back to top-level. Somebody should write a DWIM debugger that
- heuristically determines if your error is "interesting"... :-)
-
- Actually, this isn't far-fetched; the condition system makes it
- possible to customize the behavior of MIT Scheme by classifying errors
- in different ways. It's easy to distinguish different kinds of
- errors, and to abort out of some while debugging others. This
- distinction can also take context into account since every error
- carries a continuation that describes the error, and programs can
- examine continuations by means of the "stack parser" (essentially a
- programmatic debugger).
-
- What's difficult is deciding what kinds of errors you're interested
- in. Particularly since DWIM probably isn't DWYM.
-