Re: Tramp 2.1.9

[Michael Albinus]

Michael Albinus <address@hidden> writes:

Hi Pete,

>> Your plinkx patch was missing a bit.  I think there should be an entry
>> for tramp-set-completion-function in the second (eval-after-load
>> "tramp").
>
> I'm aware of this. I haven't done it yet, because I'm still thinking
> about how to exploit plinkx at best. One option would be to allow
> _only_ PuTTY sessions here, i.e. calling "plink -load sessionname"
> (instead of hostname). Consequently, auto-completion would need to
> read PuTTY session names from the Windows registry - a new function.

I've committed to CVS a change which implements this behaviour. Since
I'm not a regular Windows user, it might be helpful if you (and
everybody else :-) could test it a little bit.

Especially, I'm not confident about password-end-of-line
handling. plink and plink1 need two additional arbitrary
characters. Using a PuTTY session doesn't seem to need it. But again,
I'm not a trustworthy tester at all for Windows related software ...

Best regards, Michael.