vsuharnikov@gmail.com writes:

> Steps for reproduce:
> * start emacs: emacs -Q
> * create an empty python buffer: C-x b foo.py RET
> * turn on python mode: M-x python-mode RET
> * start interpreter: C-c C-z (python-shell-switch-to-shell) and answer the 
> prompts (I pressed RET n)
> * switch back to python buffer: C-x o (other-window)
> * type "from os import " (there is a space after "import")
> * press C-M-i (completion-at-point)
>
> Observed behaviour: "import " is inserted in the buffer. The buffer now 
> contains "from os import import ".
> Expected behaviour: completion-at-point offers names from the os module 
> as completion candidates.

I'm unable to reproduce this in Emacs 26.1-28, so I'm going to go ahead
that this has been fixed in the years since this was reported, and I'm
closing this bug report.  If you can still see this problem, please
respond to the debbugs address, and we'll reopen.

However, there seems to be a regression between 25.1 and 26.1
somewhere -- `C-M-i' doesn't give a completion at all now, and in Emacs
25.1 it used to say: