I'm trying to select a CONTINUE restart that offers to modify a place from within the slime debugger but, alas, this doesn't work. Even so I get prompted for a new value (in the REPL, not in the debugger) the evaluation gets aborted. Here's a (primitive) test case:
CL> (defparameter x 42) X CL> (assert nil (X) "Something fishy") Something fishy [Condition of type SIMPLE-ERROR]
Restarts: 0: [CONTINUE] Retry assertion with new value for X. 1: [ABORT] Return to SLIME's top level. 2: [ABORT] Exit debugger, returning to top level.
Backtrace: 0: (SB-KERNEL:ASSERT-ERROR NIL (X) "Something fishy") 1: (NIL)
(user enters '0' or 'C' here ...)
The old value of X is 42. Do you want to supply a new value? (y or n) ; Evaluation aborted. CL>
The same test works from the lisp REPL. My system:
CL> (lisp-implementation-type) "SBCL" CL> (lisp-implementation-version) "1.0.10.debian"
Slime is a current CVS checkout.
TIA Ralf Mattes
Ralf Mattes rm@mh-freiburg.de writes:
I'm trying to select a CONTINUE restart that offers to modify a place from within the slime debugger but, alas, this doesn't work. Even so I get prompted for a new value (in the REPL, not in the debugger) the evaluation gets aborted. Here's a (primitive) test case:
This is due to a stupid regression that I introduced a few days back, as has been pointed out by Helmut. It seems that he fixed it already in CVS. (Update again from CVS and make sure that the last ChangeLog entry is from 21th January.)
Sorry about that! (As punishment for not having run the test suite before checking in, I'll try to fix the test cases that have been failing since quite some time as soon as time permits.)
-T.
On Mon, 21 Jan 2008 22:46:12 +0100, Tobias C. Rittweiler wrote:
Ralf Mattes rm@mh-freiburg.de writes:
I'm trying to select a CONTINUE restart that offers to modify a place from within the slime debugger but, alas, this doesn't work. Even so I get prompted for a new value (in the REPL, not in the debugger) the evaluation gets aborted. Here's a (primitive) test case:
This is due to a stupid regression that I introduced a few days back, as has been pointed out by Helmut. It seems that he fixed it already in CVS. (Update again from CVS and make sure that the last ChangeLog entry is from 21th January.)
Ok, I must have missed that checkin. Hmm. Works fine now after the update. Thanks!
Sorry about that! (As punishment for not having run the test suite before checking in, I'll try to fix the test cases that have been failing since quite some time as soon as time permits.)
Hey, since when can the delinquent choose the punishment .-)
Cheers, RalfD
-T.
-
Ralf Mattes -
Tobias C. Rittweiler