gh-138641: prevent accessing non-existing curframe in pdb commands

[yihong0618]

breakpoint() in pdb the self.curframe will be None
and run something will cause crash

this patch fix it. but not sure if it is worth. if not will close.

• Issue: pdb commands may try to access non-existing frames #138641

[picnixz]

If you're not sure if your patch is correct yet or not, let's put it as a draft first.

[picnixz]

There are other places where curframe is used such as completedefault. Please check all the places carefully.

[bedevere-app]

A Python core developer has requested some changes be made to your pull request before we can consider merging it. If you could please address their requests along with any other requests in other reviews from core developers that would be appreciated.

Once you have made the requested changes, please leave a comment on this pull request containing the phrase I have made the requested changes; please review again. I will then notify any core developers who have left a review that you're ready for them to take another look at this pull request.

And if you don't make the requested changes, you will be poked with soft cushions!

[yihong0618]

If you're not sure if your patch is correct yet or not, let's put it as a draft first.

copy will change and open it

[yihong0618]

There are other places where curframe is used such as completedefault. Please check all the places carefully.

copy will check

[yihong0618]

There are other places where curframe is used such as completedefault. Please check all the places carefully.

check something others seems not raise error, I will left them

[picnixz]

others seems not raise error, I will left them

completedefault appears to be handled, so let's add an assert self.curframe at the beginning of the function. However, the following raise:

>>> breakpoint()
> <python-input-0>(1)<module>()
(Pdb) breakpoint()
(Pdb) debug 1

>>> breakpoint()
> <python-input-0>(1)<module>()
(Pdb) breakpoint()
(Pdb) break 1

And defaultFile may need to be patched as well because do_break() calls defaultFile(). The exceptions in do_break are silenced so there shouldn't be a need for patching them (there are bare except). However you need to patch do_debug and do_break.

[yihong0618]

others seems not raise error, I will left them

completedefault appears to be handled, so let's add an assert self.curframe at the beginning of the function. However, the following raise:

>>> breakpoint()
> <python-input-0>(1)<module>()
(Pdb) breakpoint()
(Pdb) debug 1

>>> breakpoint()
> <python-input-0>(1)<module>()
(Pdb) breakpoint()
(Pdb) break 1

And defaultFile may need to be patched as well because do_break() calls defaultFile(). The exceptions in do_break are silenced so there shouldn't be a need for patching them (there are bare except). However you need to patch do_debug and do_break.

fixed and failed test seems not releate

[yihong0618]

careful with another case
does it need to fix here?

Python 3.15.0a0 (heads/hy/inside_pdb_breakpoints_can_crash-dirty:cb73cb0750, Sep  8 2025, 19:02:2) [Clang 17.0.0 (clang-1700.0.13.5)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> breakpoint()
> <python-input-0>(1)<module>()
(Pdb) break 1
End of file
(Pdb) breakpoint()
(Pdb) breakpoint()
*** No current frame.
(Pdb) break 1
*** No current frame.
Traceback (most recent call last):
  File "<python-input-0>", line 1, in <module>
    breakpoint()
    ~~~~~~~~~~^^
  File "/Users/yihong/repos/cpython/Lib/bdb.py", line 94, in wrapper
    ret = func(frame, *args)
  File "/Users/yihong/repos/cpython/Lib/bdb.py", line 154, in opcode_callback
    frame.f_trace(frame, 'opcode', None)
    ~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/yihong/repos/cpython/Lib/bdb.py", line 289, in trace_dispatch
    return self.dispatch_opcode(frame, arg)
           ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^
  File "/Users/yihong/repos/cpython/Lib/bdb.py", line 406, in dispatch_opcode
    self.user_opcode(frame)
    ~~~~~~~~~~~~~~~~^^^^^^^
  File "/Users/yihong/repos/cpython/Lib/pdb.py", line 547, in user_line
    self.interaction(frame, None)
    ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
  File "/Users/yihong/repos/cpython/Lib/pdb.py", line 736, in interaction
    self._cmdloop()
    ~~~~~~~~~~~~~^^
  File "/Users/yihong/repos/cpython/Lib/pdb.py", line 601, in _cmdloop
    self.cmdloop()
    ~~~~~~~~~~~~^^
  File "/Users/yihong/repos/cpython/Lib/cmd.py", line 149, in cmdloop
    stop = self.onecmd(line)
  File "/Users/yihong/repos/cpython/Lib/pdb.py", line 1041, in onecmd
    return cmd.Cmd.onecmd(self, line)
           ~~~~~~~~~~~~~~^^^^^^^^^^^^
  File "/Users/yihong/repos/cpython/Lib/cmd.py", line 227, in onecmd
    return func(arg)
  File "/Users/yihong/repos/cpython/Lib/pdb.py", line 1399, in do_break
    filename = self.canonic(filename)
  File "/Users/yihong/repos/cpython/Lib/bdb.py", line 221, in canonic
    if filename == "<" + filename[1:-1] + ">":
                         ~~~~~~~~^^^^^^
TypeError: 'NoneType' object is not subscriptable

[yihong0618]

seems can into a function

logic has been changed as per Tian's suggestions

[picnixz]

I'll bail from shepherding this one as Tian's is the best qualified here. Sorry for the un-necessary work that I made you do! (and thank you for your sponsor ❤️)

[yihong0618]

I'll bail from shepherding this one as Tian's is the best qualified here. Sorry for the un-necessary work that I made you do! (and thank you for your sponsor ❤️)

no need~, my English is bad, sorry for that,
and I learned a lot from this.