Re: partial-completion-mode and eval

[Chong Yidong]

A Soare <address@hidden> writes:

> 1. Emacs -Q
>
> 2. switch on partial completion mode.
>
> 3. M-:
>
> Eval: (fo-ch <M-TAB>
>
> => (fo-char and the point position after fo.
>
> Now press
>
> Eval: (forw-char <M-TAB>
>
> => Eval: (forward--char
>
> There are 2 -- instead of 1.

After looking at this again, I think this is a generic behavior of
PC-lisp-complete-symbol in partial completion mode; it happens even
outside the minibuffer, and the logic is that PC-lisp-complete-symbol
completes the part of the symbol before point.  It's arguably true
that the alternative behavior you suggested is better, but I think we
should leave this alone, at least for now.