You are doing god’s work on a time-intensive computation, but your final step errors out :S Are you doomed to start everything from zero, and wait again for this long process? No! Find out.

I show this with Emacs and Slime, then with the Lem editor (ready-to-use for CL, works with many more languages thanks to its LSP client).

(This video is so cool :D Sound on)

We use the built-in Common Lisp interactive debugger that lets us restart one precise frame from the call stack. Once you get the debugger:

  • press “v” on a frame to go to the corresponding source file
  • fix the buggy function, compile it again (C-c C-c in Slime)
  • come back to the debugger, place the point on the frame you want to try again, press “r” (sldb-restart-frame, see your editor’s menu)
  • see everything succeed. You did not restart everything from zero.
  • if you don’t see some function calls and variables in the debugger, compile your code with max debug settings (C-u C-c C-c in Slime).

For more:

Other videos:

Hope you learned something!


Debugging a complex stacktrace in Python VS Common Lisp (unknown author)