How to continue to the next loop iteration in Python PDB?

Question

Given this sample code:

import pdb

for i in range(10):
  pdb.set_trace()
  print(str(i))

When I get the prompt from PDB, how can I skip an iteration of the loop, with the continue loop control statement, when it's also used by PDB, to continue code execution?

Answer 1

You cannot use continue because new statements in the debugger need to be complete and valid without any other context; continue must be given inside a loop construct when being compiled. As such using !continue (with the ! to prevent pdb from interpreting the command) cannot be used even if the debugger is processing a loop construct.

You can use the j[ump] command, provided you have a later statement to jump to. If your loop is empty after the statements you wanted to jump over, you can only 'rewind':

$ bin/python test.py
> /.../test.py(5)<module>()
-> print(str(i))
(Pdb) l
  1     import pdb
  2     
  3     for i in range(10):
  4         pdb.set_trace()
  5  ->     print(str(i))
  6     
[EOF]
(Pdb) j 3
> /.../test.py(3)<module>()
-> for i in range(10):

j 3 jumped to line 3, not skipping anything; line 3 will be re-executed including setting up the range(). You could jump to line 4, but then the for loop doesn't advance.

You'd need to add another statement at the end of the loop to jump to for Python to continue from. That statement can be a print() or a pass or anything else that doesn't have to alter your state. You could even use continue as the last statement. I used i:

for i in range(10):
    pdb.set_trace()
    print(str(i))
    i  # only here to jump to.

Demo:

$ bin/python test.py
> /.../test.py(5)<module>()
-> print(str(i))
(Pdb) l
  1     import pdb
  2     
  3     for i in range(10):
  4         pdb.set_trace()
  5  ->     print(str(i))
  6         i  # only here to jump to.
  7     
[EOF]
(Pdb) j 6
> /.../test.py(6)<module>()
-> i  # only here to jump to.
(Pdb) c
> /.../test.py(4)<module>()
-> pdb.set_trace()
(Pdb) s
> /.../test.py(5)<module>()
-> print(str(i))
(Pdb) j 6
> /.../test.py(6)<module>()
-> i  # only here to jump to.
(Pdb) i
1
(Pdb) c
> /.../test.py(4)<module>()
-> pdb.set_trace()
(Pdb) s
> /.../test.py(5)<module>()
-> print(str(i))
(Pdb) j 6
> /.../test.py(6)<module>()
-> i  # only here to jump to.
(Pdb) i
2

From Debugger Commands:

j(ump) lineno
Set the next line that will be executed. Only available in the bottom-most frame. This lets you jump back and execute code again, or jump forward to skip code that you don’t want to run.

It should be noted that not all jumps are allowed — for instance it is not possible to jump into the middle of a for loop or out of a finally clause.

Answer 2

Sounds like a strange thing to want to do. You should be able to use the jump command though. You'll probably need to add a pass statement at the end of your for loop so you can jump to the end of the loop. If you're not sure of the line numbers of your code then you can use ll to find out the line numbers of your loop.

> c:\run.py(5)<module>()
-> print(i)
(Pdb) ll
  1     import pdb
  2     
  3     for i in range(10):
  4         pdb.set_trace()
  5  ->     print(i)
  6         pass
(Pdb) j 6
> c:\run.py(6)<module>()
-> pass
(Pdb) c
> c:\python\run.py(4)<module>()
-> pdb.set_trace()
(Pdb) c
1
> c:\python\run.py(5)<module>()
-> print(i)

It's worth noting that jumping to the for line will restart the loop.