bug-gnu-emacs
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

bug#74626: 30.0.91; mini buffer password prompt mismatches shell buffer'


From: Michael Albinus
Subject: bug#74626: 30.0.91; mini buffer password prompt mismatches shell buffer's
Date: Sun, 01 Dec 2024 16:44:37 +0100
User-agent: Gnus/5.13 (Gnus v5.13)

Van Ly via "Bug reports for GNU Emacs, the Swiss army knife of text
editors" <bug-gnu-emacs@gnu.org> writes:

> Hello,

Hi,

> When using ssh to connect to a machine, the password prompt is expected
> to be identical in the shell buffer and mini buffer.
>
> The mini buffer's prompt seems to make use of only the last character.
>
> See
>
>   http://sdf.org/~van.ly/img/emacs-30-0-91-password-prompt.webp
>
> The shell buffer's password prompt shows
>
>   van@192.168.1.144's password:
>
> The mini buffer shows
>
>   4's password:

Thanks for the report.

I can reproduce the problem. In fact, there are two issues:

- comint-password-prompt-regexp uses "'s" in order to search for a password
  prompt. That's fine for password prompts like "van@192.168.1.144's
  password:", however, everything before this "'s" is ignored when the
  minibuffer prompt is assembled.

- The string the password prompt is looked for is "van@192.168.1.144's
  password:". In comint-watch-for-password-prompt, the "" is removed,
  but while searching for the password prompt. As a result,
  (match-string 0 string) returns a wrong substring (just one position
  left).

The problem does not happen with Emacs 29. The culprit seems to be
commit 1a55e957ae57, which has changed the implementation of
comint-watch-for-password-prompt. The appended patch fixes this for me
in Emacs 30. Could you, please, check?

Jim, do you have a comment?

Best regards, Michael.

Attachment: txtV1nNZibbsv.txt
Description: Text Data


reply via email to

[Prev in Thread] Current Thread [Next in Thread]