Re: Using the MIT Scheme Stepper (debugging), success

[Nicholas Papadonis]

I had some success with the Stepper.

(define (test x)

(car x)

        (cdr x))

(test ‘(a b c))

Place cursor at the beginning of (test ‘(a b c))

Then:

M-x step-defun

Then:

press space to step

Seems like a useful feature!

Nick

On Apr 3, 2020, at 8:18 AM, Nicholas Papadonis <address@hidden> wrote:

I opened runtime/ystep.scm and edwin/eystep.scm and used eval-current-buffer.  There are many unbound variables and the debugger is opened.  Also, what function should be applied to start stepping.

I’m in the middle of SICP so any guidance on using this interface is appreciated.

Thanks

On Mar 27, 2020, at 3:56 AM, Chris Hanson <address@hidden> wrote:

The debugger can’t examine what will happen in the future, because that hasn’t happened yet.

If you want to move forward, you have several options:

1. Resume the program using one of the restarts that are shown when you stop.
2. Evaluate subexpressions in the debugger to see what they do.
3. Run the stepper to evaluate expressions one step at a time.

The stepper isn’t documented but the source files are runtime/ystep.scm for the basic stepper and edwin/eystep.scm for the Edwin stepper interface.  Caveats: the stepper works only for interpreted code, not compiled code, so you can’t step into compiled code to see what it’s doing.

On Mar 26, 2020, 5:20 PM -0700, Nicholas Papadonis <address@hidden>, wrote:

When I insert (bkpt) in the code it launches the debugger, it appears the debugger can move backwards in subproblems/reductions, however cannot move forward past the (bkpt).

Is there a way to evaluate expressions forward past the break point?

I did not see an option in the debugger help. 

Thanks