[plt-scheme] Proposal for s-expr tracing format

From: Carl Eastlund (cce at ccs.neu.edu)
Date: Mon Aug 13 10:08:37 EDT 2007

On 8/13/07, Richard Cobbe <cobbe at ccs.neu.edu> wrote:
> On Mon, Aug 13, 2007 at 09:34:27AM -0400, Matthias Felleisen wrote:
> > Question is whether you do want the parens closed. I like John's idea that
> > an open paren shows control transfer.
>
> John's suggestion has a certain appeal -- the missing parenthesis just
> "feels right" to represent an abnormal exit from a function.
>
> I'm not sure it's the right thing from a UI perspective, though.  Perhaps
> the OP's proposed output makes this objection irrelevant, but I'd worry
> that the missing close paren will get lost in the rest of the output --
> especially since we keep telling our students that they shouldn't pay
> attention to the parentheses!

With hanging parens, it's possible to really mess up the s-expression
structure.  By calling (and aborting) a function many times, or
invoking a saved continuation to finish a called function many times,
one can create arbitrary parentheses, none of which actually
correspond to their matching open/close.  It's probably better to
simply record any escape/resume via continuations as the close or open
of a new set of parentheses.

-- 
Carl Eastlund


Posted on the users mailing list.