[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
bug#34489: 25.2; pdb fails if directory contains '++'
From: |
Stefan Kangas |
Subject: |
bug#34489: 25.2; pdb fails if directory contains '++' |
Date: |
Sun, 30 Jun 2019 18:13:19 +0200 |
a.soroa@ehu.eus (a.soroa) writes:
> 'pdb' command does not work if the directory where the python script is
> contains the characters '++'. Here is a receipt
>
> $ mkdir /tmp/foo++
> $ cd /tmp/foo++
> $ echo "print(10)" > pp.py
> $ emacs -Q
> M-x pdb
> Run pdb (like this): python3 -m pdb pp.py
I am able to reproduce this on current master.
The expected behaviour is to see TWO buffers in separate windows,
first "*gud-pp.py" with the following content:
Current directory is /tmp/foo++/
> /tmp/foo++/pp.py(1)<module>()
-> print(10)
(Pdb)
And the second a buffer visiting "pp.py".
The symptom is that you only see ONE buffer "*gud-pp.py*" with the
following content:
Current directory is /tmp/foo++/
The bug (half) goes away as soon as I try to debug `gud-common-init'
using M-x edebug-defun, but only when stepping through manually using
"n" (and not when hitting "c"). Without this instrumentation, I get
the bug as above. With it, I see the expected output in the
"*gud-pp.py*" buffer (but *not* the expected "pp.py" buffer).
I tried adding strategic calls to sit-for in gud-common-init and could
get the same result as with M-x edebug-defun.
Does anyone have any ideas what could cause this?
Thanks,
Stefan Kangas
- bug#34489: 25.2; pdb fails if directory contains '++',
Stefan Kangas <=