GNU bug report logs - #6584
23.2; M-p (comint-previous-input) breaks intermittently in telnet-mode.

Previous Next

Full log

View this message in rfc822 format

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: David Keegan <david.keegan <at> shenick.com>
Cc: 6584 <at> debbugs.gnu.org
Subject: bug#6584: 23.2; M-p (comint-previous-input) breaks intermittently in telnet-mode.
Date: Tue, 08 Dec 2020 18:45:32 +0100

David Keegan <david.keegan <at> shenick.com> writes:

> 1. M-p (comint-previous-input) breaks. Instead of retrieving
> the previous command to the RHS of the prompt it places it
> on the LHS of the prompt where it is useless.
>
> This is caused by function
> telnet-check-software-type-initialize which sets
> comint-prompt-regexp inappropriately to "^[^*>\n]*[*>] *"
> when it sees "its" in the remote host output.

(This bug report unfortunately got no response at the time.)

I'm unable to reproduce this in Emacs 28.

> 2. Spurious prompts for a password if "passw" happens to
> occur in the remote host output.

This bug is still present on the current trunk.  To reproduce

emacs -Q
M-x rsh RET hostname RET
touch /tmp/passw
ls /tmp/

Emacs will now query you for a password in the minibuffer.

And things are kinda odd all over.  If I have:

[larsi <at> stories ~]$

and then

[larsi <at> stories ~]$ cd RET

I get

[larsi <at> stories ~]$ [larsi <at> stories ~]$ 

this is the case with all commands that don't output anything.

My guess is that nobody has used the rsh command in a while?  At least
not over ssh?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.