A recursive debugging REPL exposes a number of other meta-commands that inspect the state of the computation at the time of the error. These commands allow you to
See Debug Commands, for documentation of the individual commands. This section aims to give more of a walkthrough of a typical debugging session.
First, we’re going to need a good error. Let’s try to macroexpand the
expression (unquote foo), outside of a quasiquote form,
and see how the macroexpander reports this error.
scheme@(guile-user)> (macroexpand '(unquote foo)) ERROR: In procedure macroexpand: ERROR: unquote: expression not valid outside of quasiquote in (unquote foo) Entering a new prompt. Type `,bt' for a backtrace or `,q' to continue. scheme@(guile-user) [1]>
The backtrace command, which can also be invoked as bt,
displays the call stack (aka backtrace) at the point where the debugger
was entered:
scheme@(guile-user) [1]> ,bt
In ice-9/psyntax.scm:
1130:21 3 (chi-top (unquote foo) () ((top)) e (eval) (hygiene #))
1071:30 2 (syntax-type (unquote foo) () ((top)) #f #f (# #) #f)
1368:28 1 (chi-macro #<procedure de9360 at ice-9/psyntax.scm...> ...)
In unknown file:
0 (scm-error syntax-error macroexpand "~a: ~a in ~a" # #f)
A call stack consists of a sequence of stack frames, with each frame describing one procedure which is waiting to do something with the values returned by another. Here we see that there are four frames on the stack.
Note that macroexpand is not on the stack – it must have made a
tail call to chi-top, as indeed we would find if we searched
ice-9/psyntax.scm for its definition.
When you enter the debugger, the innermost frame is selected, which
means that the commands for getting information about the “current”
frame, or for evaluating expressions in the context of the current
frame, will do so by default with respect to the innermost frame. To
select a different frame, so that these operations will apply to it
instead, use the up, down and frame commands like
this:
scheme@(guile-user) [1]> ,up In ice-9/psyntax.scm: 1368:28 1 (chi-macro #<procedure de9360 at ice-9/psyntax.scm...> ...) scheme@(guile-user) [1]> ,frame 3 In ice-9/psyntax.scm: 1130:21 3 (chi-top (unquote foo) () ((top)) e (eval) (hygiene #)) scheme@(guile-user) [1]> ,down In ice-9/psyntax.scm: 1071:30 2 (syntax-type (unquote foo) () ((top)) #f #f (# #) #f)
Perhaps we’re interested in what’s going on in frame 2, so we take a look at its local variables:
scheme@(guile-user) [1]> ,locals Local variables: $1 = e = (unquote foo) $2 = r = () $3 = w = ((top)) $4 = s = #f $5 = rib = #f $6 = mod = (hygiene guile-user) $7 = for-car? = #f $8 = first = unquote $9 = ftype = macro $10 = fval = #<procedure de9360 at ice-9/psyntax.scm:2817:2 (x)> $11 = fe = unquote $12 = fw = ((top)) $13 = fs = #f $14 = fmod = (hygiene guile-user)
All of the values are accessible by their value-history names
($n):
scheme@(guile-user) [1]> $10 $15 = #<procedure de9360 at ice-9/psyntax.scm:2817:2 (x)>
We can even invoke the procedure at the REPL directly:
scheme@(guile-user) [1]> ($10 'not-going-to-work) ERROR: In procedure macroexpand: ERROR: source expression failed to match any pattern in not-going-to-work Entering a new prompt. Type `,bt' for a backtrace or `,q' to continue.
Well at this point we’ve caused an error within an error. Let’s just quit back to the top level:
scheme@(guile-user) [2]> ,q scheme@(guile-user) [1]> ,q scheme@(guile-user)>
Finally, as a word to the wise: hackers close their REPL prompts with C-d.