GNU bug report logs -
#75746
31.0.50; Can't get password promote from plink 0.82
Previous Next
Reported by: 孙曙光(Sun Shuguang) <shuguang.sun <at> zaiming.com>
Date: Wed, 22 Jan 2025 07:49:02 UTC
Severity: normal
Found in version 31.0.50
Fixed in version 30.2
Done: Michael Albinus <michael.albinus <at> gmx.de>
Bug is archived. No further changes may be made.
To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 75746 in the body.
You can then email your comments to 75746 AT debbugs.gnu.org in the normal way.
Toggle the display of automated, internal messages from the tracker.
Report forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Wed, 22 Jan 2025 07:49:02 GMT)
Full text and
rfc822 format available.
Acknowledgement sent
to
孙曙光(Sun Shuguang) <shuguang.sun <at> zaiming.com>
:
New bug report received and forwarded. Copy sent to
bug-gnu-emacs <at> gnu.org
.
(Wed, 22 Jan 2025 07:49:03 GMT)
Full text and
rfc822 format available.
Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):
Dear,
Emacs (31.0.50, UCRT) can't get the promote for password (and user) from plink 0.82 and pre-release 0.83 in Windows. So that Tramp using plink from windows will not work.
- Emacs -q
- shell (cmd)
- plink -l username -P 22 -ssh -t 192.168.x.x " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh " && exit || exit
The shell echo the username and wait and wait.
- plink -P 22 -ssh -t 192.168.x.x " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh " && exit || exit
The shell has no echo and wait and wait.
In Tramp, it falls into the loop of waiting for promote of password.
Best Regards,
Shuguang
Below is the system info:
In GNU Emacs 31.0.50 (build 1, x86_64-w64-mingw32)
Repository revision: 299a1fbef886dba654f58a0c55b373f8b43d383b
Repository branch: HEAD
Windowing system distributor 'Microsoft Corp.', version 10.0.26100
System Description: Microsoft Windows 10 Home China (v10.0.2009.26100.2894)
Configured using:
'configure
--prefix=/d/a/emacs-build/emacs-build/pkg/299a1fb-ucrt-x86_64
'CFLAGS=-O2 -fno-semantic-interposition -floop-parallelize-all
-ftree-parallelize-loops=4 -g3 ' --disable-build-details --without-dbus
--enable-link-time-optimization --with-compress-install
--with-small-ja-dic --with-gif --with-gnutls --with-harfbuzz
--with-jpeg --with-json --with-lcms2 --with-native-compilation
--with-png --with-rsvg --with-tree-sitter --with-xml2 --with-xpm
--with-zlib --without-cairo --without-mps --without-tiff'
Configured features:
ACL GIF GMP GNUTLS HARFBUZZ JPEG LCMS2 LIBXML2 MODULES NATIVE_COMP
NOTIFY W32NOTIFY PDUMPER PNG RSVG SOUND SQLITE3 THREADS
TOOLKIT_SCROLL_BARS TREE_SITTER WEBP XPM ZLIB
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Sun, 26 Jan 2025 09:01:02 GMT)
Full text and
rfc822 format available.
Message #8 received at 75746 <at> debbugs.gnu.org (full text, mbox):
(Sun Shuguang) <shuguang.sun <at> zaiming.com> writes:
> Dear,
Hi Shuguang,
> Emacs (31.0.50, UCRT) can't get the promote for password (and user) from plink 0.82 and pre-release 0.83 in Windows. So that Tramp using plink from windows will not work.
>
> - Emacs -q
> - shell (cmd)
> - plink -l username -P 22 -ssh -t 192.168.x.x " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh " && exit || exit
> The shell echo the username and wait and wait.
> - plink -P 22 -ssh -t 192.168.x.x " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh " && exit || exit
> The shell has no echo and wait and wait.
>
> In Tramp, it falls into the loop of waiting for promote of password.
Could you pleas show the output of the two commands in a cmd.exe window?
With and without the "&& exit || exit" tail.
> Best Regards,
> Shuguang
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Mon, 27 Jan 2025 05:23:03 GMT)
Full text and
rfc822 format available.
Message #11 received at 75746 <at> debbugs.gnu.org (full text, mbox):
>
> Hi Shuguang,
>
> > Emacs (31.0.50, UCRT) can't get the promote for password (and user)
> from plink 0.82 and pre-release 0.83 in Windows. So that Tramp using plink
> from windows will not work.
> >
> > - Emacs -q
> > - shell (cmd)
> > - plink -l username -P 22 -ssh -t 192.168.x.x " env 'TERM=dumb'
> 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh " && exit || exit
> > The shell echo the username and wait and wait.
> > - plink -P 22 -ssh -t 192.168.x.x " env 'TERM=dumb'
> 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh " && exit || exit
> > The shell has no echo and wait and wait.
> >
> > In Tramp, it falls into the loop of waiting for promote of password.
>
> Could you pleas show the output of the two commands in a cmd.exe
> window?
> With and without the "&& exit || exit" tail.
>
For command with `-l username`, the output is ` username <at> 192.168.x.x's password:`.
For command without `-l username` the output is ` login as:`.
It is the same with or without the "&& exit || exit" tail.
> > Best Regards,
> > Shuguang
>
> Best regards, Michael.
BR
Shuguang
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Mon, 27 Jan 2025 08:00:02 GMT)
Full text and
rfc822 format available.
Message #14 received at 75746 <at> debbugs.gnu.org (full text, mbox):
"孙曙光(Sun Shuguang)" <shuguang.sun <at> zaiming.com> writes:
Hi Shuguang,
>> Could you pleas show the output of the two commands in a cmd.exe
>> window?
>> With and without the "&& exit || exit" tail.
>
> For command with `-l username`, the output is ` username <at> 192.168.x.x's password:`.
> For command without `-l username` the output is ` login as:`.
>
> It is the same with or without the "&& exit || exit" tail.
Thanks, this looks proper. Could you please call Emacs like "emacs -Q
--eval '(setq tramp-verbose 10)'"? Try to connect via plink. If it hangs
or runs into the timeout, send the resulting Tramp debug buffer "*debug
tramp/plink 192.168.x.x*" or "*debug tramp/plink username <at> 192.168.x.x*".
> BR
> Shuguang
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Wed, 29 Jan 2025 15:21:02 GMT)
Full text and
rfc822 format available.
Message #17 received at 75746 <at> debbugs.gnu.org (full text, mbox):
[Message part 1 (text/plain, inline)]
> >> Could you pleas show the output of the two commands in a cmd.exe
> >> window?
> >> With and without the "&& exit || exit" tail.
> >
> > For command with `-l username`, the output is ` username <at> 192.168.x.x's
> password:`.
> > For command without `-l username` the output is ` login as:`.
> >
> > It is the same with or without the "&& exit || exit" tail.
>
> Thanks, this looks proper. Could you please call Emacs like "emacs -Q --eval
> '(setq tramp-verbose 10)'"? Try to connect via plink. If it hangs or runs into
> the timeout, send the resulting Tramp debug buffer "*debug tramp/plink
> 192.168.x.x*" or "*debug tramp/plink username <at> 192.168.x.x*".
Please find the attachment. I have deleted some repeated part to reduce the file size.
>
> Best regards, Michael.
Best Regards,
Shuguang
[tramp_test.txt (text/plain, attachment)]
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Thu, 30 Jan 2025 12:57:01 GMT)
Full text and
rfc822 format available.
Message #20 received at 75746 <at> debbugs.gnu.org (full text, mbox):
"孙曙光(Sun Shuguang)" <shuguang.sun <at> zaiming.com> writes:
Hi Shuguang,
>> Thanks, this looks proper. Could you please call Emacs like "emacs -Q --eval
>> '(setq tramp-verbose 10)'"? Try to connect via plink. If it hangs or runs into
>> the timeout, send the resulting Tramp debug buffer "*debug tramp/plink
>> 192.168.x.x*" or "*debug tramp/plink username <at> 192.168.x.x*".
>
> Please find the attachment. I have deleted some repeated part to reduce the file size.
Hmm. I've installed PuTTY 0.82 on my Windows 11 VM, running Emacs
29.4. The same problem happens.
I have no idea (yet), how to solve it, although I've played with
different changes the last hours.
What looks suspicious are the PuTTY Release Notes
<https://www.chiark.greenend.org.uk/~sgtatham/putty/changes.html>:
--8<---------------cut here---------------start------------->8---
These features are new in 0.82 (released 2024-11-27):
Major refactoring of Unicode handling to allow the use of 'foreign' Unicode characters, i.e. outside the system's configured default character set / code page. Not yet complete, but the following things work:
Windows console: usernames and passwords entered interactively into PSCP, PSFTP and Plink can use foreign characters.
Windows command line: usernames, remote commands, and filenames entered via command-line options can use foreign characters.
PuTTY's own terminal (on Windows and Unix): even if it's not configured into UTF-8 mode for the main session, interactive usernames and passwords can use foreign characters.
Unicode version update: all character analysis is updated to Unicode 16.0.0.
Unicode terminal rendering: national and regional flags are now understood by PuTTY's terminal emulator. (However, correct display of those flags will depend on fonts and operating system.)
The Event Log mentions the local address and port number of the outgoing connection socket.
Bracketed paste mode can now be turned off in the Terminal > Features panel.
Unix Pageant: new --foreground mode for running as a subprocess.
Bug fix: the 'border width' configuration option is now honoured even when the window is maximised.
Bug fix: SHA-2 based RSA signatures are now sent with correct zero padding.
Bug fix: terminal wrap mishandling caused occasional incorrect redraws in curses-based applications.
Bug fix: Alt + function key in "Xterm 216+" mode sent a spurious extra escape character.
--8<---------------cut here---------------end--------------->8---
I suspect this is related to the changed behaviour. I have no idea, what
I shall do in Tramp.
FTR, PuTTY 0.82 on my Fedora 41 laptop works as expected.
For the time being, I recommend to fall back to PuTTY 0.81, or use the
MS Windoes ssh client.
> Best Regards,
> Shuguang
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Thu, 30 Jan 2025 15:27:02 GMT)
Full text and
rfc822 format available.
Message #23 received at 75746 <at> debbugs.gnu.org (full text, mbox):
> Cc: "75746 <at> debbugs.gnu.org" <75746 <at> debbugs.gnu.org>,
> Shuguang Sun <shuguang79 <at> qq.com>
> Date: Thu, 30 Jan 2025 13:55:45 +0100
> From: Michael Albinus via "Bug reports for GNU Emacs,
> the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org>
>
> "孙曙光(Sun Shuguang)" <shuguang.sun <at> zaiming.com> writes:
>
> Hi Shuguang,
>
> >> Thanks, this looks proper. Could you please call Emacs like "emacs -Q --eval
> >> '(setq tramp-verbose 10)'"? Try to connect via plink. If it hangs or runs into
> >> the timeout, send the resulting Tramp debug buffer "*debug tramp/plink
> >> 192.168.x.x*" or "*debug tramp/plink username <at> 192.168.x.x*".
> >
> > Please find the attachment. I have deleted some repeated part to reduce the file size.
>
> Hmm. I've installed PuTTY 0.82 on my Windows 11 VM, running Emacs
> 29.4. The same problem happens.
>
> I have no idea (yet), how to solve it, although I've played with
> different changes the last hours.
>
> What looks suspicious are the PuTTY Release Notes
> <https://www.chiark.greenend.org.uk/~sgtatham/putty/changes.html>:
>
> --8<---------------cut here---------------start------------->8---
> These features are new in 0.82 (released 2024-11-27):
>
> Major refactoring of Unicode handling to allow the use of 'foreign' Unicode characters, i.e. outside the system's configured default character set / code page. Not yet complete, but the following things work:
> Windows console: usernames and passwords entered interactively into PSCP, PSFTP and Plink can use foreign characters.
> Windows command line: usernames, remote commands, and filenames entered via command-line options can use foreign characters.
> PuTTY's own terminal (on Windows and Unix): even if it's not configured into UTF-8 mode for the main session, interactive usernames and passwords can use foreign characters.
> Unicode version update: all character analysis is updated to Unicode 16.0.0.
> Unicode terminal rendering: national and regional flags are now understood by PuTTY's terminal emulator. (However, correct display of those flags will depend on fonts and operating system.)
> The Event Log mentions the local address and port number of the outgoing connection socket.
> Bracketed paste mode can now be turned off in the Terminal > Features panel.
> Unix Pageant: new --foreground mode for running as a subprocess.
> Bug fix: the 'border width' configuration option is now honoured even when the window is maximised.
> Bug fix: SHA-2 based RSA signatures are now sent with correct zero padding.
> Bug fix: terminal wrap mishandling caused occasional incorrect redraws in curses-based applications.
> Bug fix: Alt + function key in "Xterm 216+" mode sent a spurious extra escape character.
> --8<---------------cut here---------------end--------------->8---
>
> I suspect this is related to the changed behaviour. I have no idea, what
> I shall do in Tramp.
Does the user name or password include non-ASCII characters?
Does Tramp assume something about the encoding of stuff sent to plink
and received from plink?
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Thu, 30 Jan 2025 16:54:02 GMT)
Full text and
rfc822 format available.
Message #26 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Eli Zaretskii <eliz <at> gnu.org> writes:
Hi Eli,
> Does the user name or password include non-ASCII characters?
No, at least not in my own test. I've used albinus <at> gandalf as tartget,
which sounds kind of magic, but is just ASCII.
No password entered, because the password prompt didn't appear.
> Does Tramp assume something about the encoding of stuff sent to plink
> and received from plink?
Not that I'm aware of.
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Thu, 30 Jan 2025 19:26:02 GMT)
Full text and
rfc822 format available.
Message #29 received at 75746 <at> debbugs.gnu.org (full text, mbox):
> From: Michael Albinus <michael.albinus <at> gmx.de>
> Cc: shuguang.sun <at> zaiming.com, 75746 <at> debbugs.gnu.org, shuguang79 <at> qq.com
> Date: Thu, 30 Jan 2025 17:53:15 +0100
>
> Eli Zaretskii <eliz <at> gnu.org> writes:
>
> Hi Eli,
>
> > Does the user name or password include non-ASCII characters?
>
> No, at least not in my own test. I've used albinus <at> gandalf as tartget,
> which sounds kind of magic, but is just ASCII.
>
> No password entered, because the password prompt didn't appear.
>
> > Does Tramp assume something about the encoding of stuff sent to plink
> > and received from plink?
>
> Not that I'm aware of.
Then maybe plink now expects UTF-16 encoding of text sent to it?
Or maybe we should talk to the PuTTY developer and ask him to give us
ideas about what could be wrong and how to solve it?
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Sat, 01 Feb 2025 09:27:02 GMT)
Full text and
rfc822 format available.
Message #32 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Eli Zaretskii <eliz <at> gnu.org> writes:
Hi Eli,
>> > Does the user name or password include non-ASCII characters?
>>
>> No, at least not in my own test. I've used albinus <at> gandalf as tartget,
>> which sounds kind of magic, but is just ASCII.
>>
>> No password entered, because the password prompt didn't appear.
>>
>> > Does Tramp assume something about the encoding of stuff sent to plink
>> > and received from plink?
>>
>> Not that I'm aware of.
>
> Then maybe plink now expects UTF-16 encoding of text sent to it?
No. I've set the coding-system-argument of the shell buffer to
utf-16. The result is unreadable text.
> Or maybe we should talk to the PuTTY developer and ask him to give us
> ideas about what could be wrong and how to solve it?
Yep. But first, here is a recipe w/o Tramp, so people can reproduce it
locally. On my Windwo 11 VM, I have installed plink 0.82. If I use the
Windows console (cmd.exe), I can do the following:
--8<---------------cut here---------------start------------->8---
C:\Users\albinus>plink -ssh -t gandalf " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh -i "
The host key is not cached for this server:
gandalf (port 22)
You have no guarantee that the server is the computer you
think it is.
The server's ssh-ed25519 key fingerprint is:
ssh-ed25519 255 SHA256:C96ox1WRNvHtxKlAVJYPlAF77hiCkHPvwgD71+IpY0k
If you trust this host, enter "y" to add the key to Plink's
cache and carry on connecting.
If you want to carry on connecting just once, without adding
the key to the cache, enter "n".
If you do not trust this host, press Return to abandon the
connection.
Store key in cache? (y/n, Return cancels connection, i for more info)
--8<---------------cut here---------------end--------------->8---
This looks normal.
Using Emacs 29.4 on that machine, I set shell-file-name to
"cmd.exe". Then I do 'M-x shell', opening the Windows console in
Emacs. Here we see:
--8<---------------cut here---------------start------------->8---
c:\Program Files\Emacs>plink -ssh -t gandalf " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh -i "
plink -ssh -t gandalf " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh -i "
--8<---------------cut here---------------end--------------->8---
No output, and the program does not return.
However, if I use a non-existing host name, plink reports this:
--8<---------------cut here---------------start------------->8---
c:\Program Files\Emacs>plink -ssh -t hostnamedoesnotexist " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh -i "
plink -ssh -t hostnamedoesnotexist " env 'TERM=dumb' 'PROMPT_COMMAND=' 'PS1=#$ ' /bin/sh -i "
Unable to open connection:
Host does not exist
--8<---------------cut here---------------end--------------->8---
I'm a n00b wrt Microsoft Windows. Perhaps somebody can investigate?
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 12:59:01 GMT)
Full text and
rfc822 format available.
Message #35 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Michael Albinus <michael.albinus <at> gmx.de> writes:
Hi Eli,
>> Or maybe we should talk to the PuTTY developer and ask him to give us
>> ideas about what could be wrong and how to solve it?
>
> Yep.
The following message has been sent to the PuTTY developers
--8<---------------cut here---------------start------------->8---
From: Michael Albinus <michael.albinus <at> gmx.de>
Subject: plink ceased to show prompts since release 0.82
To: <putty <at> projects.tartarus.org>
Date: Fri, 14 Feb 2025 13:54:22 +0100 (2 minutes, 4 seconds ago)
Hi,
I am Michael Albinus, one of the contributors to Emacs. We use plink
inmternally in order to access remote hosts, and to copy/edit/delete
files on the remote host based on this.
In order to make this happen, Emacs opens internally a cmd.exe buffer. In
this buffer, a command like
plink -l albinus -ssh -t gandalf /bin/sh -i
Nothing happens, no prompt appears.
If I call the very same command in a native cmd.exe terminal, evrything
works as expected. And so it did inside Emacs prior PuTTY 0.82.
Same behavior with PuTTY 0.83. I'm using an MS Windows 11 VM for tests,
if this matters.
Best regards, Michael.
--8<---------------cut here---------------end--------------->8---
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 13:31:02 GMT)
Full text and
rfc822 format available.
Message #38 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Michael Albinus <michael.albinus <at> gmx.de> writes:
> The following message has been sent to the PuTTY developers
And here's the answer
--8<---------------cut here---------------start------------->8---
From: Jacob Nevins <jacobn+puttybugs <at> chiark.greenend.org.uk>
Subject: Re: [putty] plink ceased to show prompts since release 0.82
To: Michael Albinus <michael.albinus <at> gmx.de>
Cc: putty <at> projects.tartarus.org
Date: Fri, 14 Feb 2025 13:01:12 +0000 (1 minute, 3 seconds ago)
Reply-To: putty <at> projects.tartarus.org
Mail-Followup-To: putty <at> projects.tartarus.org, Michael Albinus
<michael.albinus <at> gmx.de>
Michael Albinus writes:
> In order to make this happen, Emacs opens internally a cmd.exe buffer. In
> this buffer, a command like
>
> --8<---------------cut here---------------start------------->8---
> plink -l albinus -ssh -t gandalf /bin/sh -i
> --8<---------------cut here---------------end--------------->8---
>
> Nothing happens, no prompt appears.
>
> If I call the very same command in a native cmd.exe terminal, evrything
> works as expected. And so it did inside Emacs prior PuTTY 0.82.
This was a deliberate behaviour change in 0.82, described in the
following link, which also describes how to restore the old behaviour.
https://www.chiark.greenend.org.uk/~sgtatham/putty/wishlist/windows-stdio-prompts.html
--8<---------------cut here---------------end--------------->8---
The proposed option works as expected.
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 14:01:01 GMT)
Full text and
rfc822 format available.
Message #41 received at 75746 <at> debbugs.gnu.org (full text, mbox):
> From: Michael Albinus <michael.albinus <at> gmx.de>
> Cc: shuguang.sun <at> zaiming.com, 75746 <at> debbugs.gnu.org, shuguang79 <at> qq.com
> Date: Fri, 14 Feb 2025 14:29:49 +0100
>
> Michael Albinus <michael.albinus <at> gmx.de> writes:
>
> > The following message has been sent to the PuTTY developers
>
> And here's the answer
>
> --8<---------------cut here---------------start------------->8---
> From: Jacob Nevins <jacobn+puttybugs <at> chiark.greenend.org.uk>
> Subject: Re: [putty] plink ceased to show prompts since release 0.82
> To: Michael Albinus <michael.albinus <at> gmx.de>
> Cc: putty <at> projects.tartarus.org
> Date: Fri, 14 Feb 2025 13:01:12 +0000 (1 minute, 3 seconds ago)
> Reply-To: putty <at> projects.tartarus.org
> Mail-Followup-To: putty <at> projects.tartarus.org, Michael Albinus
> <michael.albinus <at> gmx.de>
>
> Michael Albinus writes:
> > In order to make this happen, Emacs opens internally a cmd.exe buffer. In
> > this buffer, a command like
> >
> > --8<---------------cut here---------------start------------->8---
> > plink -l albinus -ssh -t gandalf /bin/sh -i
> > --8<---------------cut here---------------end--------------->8---
> >
> > Nothing happens, no prompt appears.
> >
> > If I call the very same command in a native cmd.exe terminal, evrything
> > works as expected. And so it did inside Emacs prior PuTTY 0.82.
>
> This was a deliberate behaviour change in 0.82, described in the
> following link, which also describes how to restore the old behaviour.
> https://www.chiark.greenend.org.uk/~sgtatham/putty/wishlist/windows-stdio-prompts.html
> --8<---------------cut here---------------end--------------->8---
>
> The proposed option works as expected.
Thanks.
I'm not sure I understand the description of the new behavior: it says
"(If there's no console involved, they'll fall back to the old
behaviour.)", but in our case, i.e. invoking PuTTY from Emacs on
Windows, there really isn't any console, since Emacs communicates with
PuTTY via a pipe. Or am I missing something?
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 14:08:02 GMT)
Full text and
rfc822 format available.
Message #44 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Eli Zaretskii <eliz <at> gnu.org> writes:
Hi,
>> This was a deliberate behaviour change in 0.82, described in the
>> following link, which also describes how to restore the old behaviour.
>> https://www.chiark.greenend.org.uk/~sgtatham/putty/wishlist/windows-stdio-prompts.html
>> --8<---------------cut here---------------end--------------->8---
>>
>> The proposed option works as expected.
>
> Thanks.
>
> I'm not sure I understand the description of the new behavior: it says
> "(If there's no console involved, they'll fall back to the old
> behaviour.)", but in our case, i.e. invoking PuTTY from Emacs on
> Windows, there really isn't any console, since Emacs communicates with
> PuTTY via a pipe. Or am I missing something?
Tramp opens a new local process, running cmdproxy. In its process
buffer, it calls plink. So it might look like a console from the PuTTY
POV.
Best regards, Michael.
Reply sent
to
Michael Albinus <michael.albinus <at> gmx.de>
:
You have taken responsibility.
(Fri, 14 Feb 2025 14:31:02 GMT)
Full text and
rfc822 format available.
Notification sent
to
孙曙光(Sun Shuguang) <shuguang.sun <at> zaiming.com>
:
bug acknowledged by developer.
(Fri, 14 Feb 2025 14:31:02 GMT)
Full text and
rfc822 format available.
Message #49 received at 75746-done <at> debbugs.gnu.org (full text, mbox):
Version: 30.2
>> If I call the very same command in a native cmd.exe terminal, evrything
>> works as expected. And so it did inside Emacs prior PuTTY 0.82.
>
> This was a deliberate behaviour change in 0.82, described in the
> following link, which also describes how to restore the old behaviour.
> https://www.chiark.greenend.org.uk/~sgtatham/putty/wishlist/windows-stdio-prompts.html
>
> The proposed option works as expected.
I've adapted the call of plink, pscp, and psftp accordingly. This
requires now at least PuTTY 0.82.
Pushed to the master branch. It will appear also with Tramp 2.7.2.2 on
GNU ELPA, later this month. Closing the bug
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 15:19:02 GMT)
Full text and
rfc822 format available.
Message #52 received at 75746 <at> debbugs.gnu.org (full text, mbox):
> From: Michael Albinus <michael.albinus <at> gmx.de>
> Cc: shuguang.sun <at> zaiming.com, 75746 <at> debbugs.gnu.org, shuguang79 <at> qq.com
> Date: Fri, 14 Feb 2025 15:07:11 +0100
>
> Eli Zaretskii <eliz <at> gnu.org> writes:
>
> > I'm not sure I understand the description of the new behavior: it says
> > "(If there's no console involved, they'll fall back to the old
> > behaviour.)", but in our case, i.e. invoking PuTTY from Emacs on
> > Windows, there really isn't any console, since Emacs communicates with
> > PuTTY via a pipe. Or am I missing something?
>
> Tramp opens a new local process, running cmdproxy.
How does Tramp start cmdproxy? If that's via make-process or
start-process, then the cmdproxy's standard input and output are
connected to a pipe.
> In its process buffer, it calls plink. So it might look like a
> console from the PuTTY POV.
I don't think so, see above.
Perhaps PuTTY considers anything that is not a regular file a console?
Otherwise, I don't understand why we see this issue.
But if you are okay with forcing the new command-line option on plink,
I don't mind. I'm just worried what will happen to older versions of
plink when they are passed this option -- if they error out, we'll
need to detect the version or ask users to configure it, which is not
so good.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 15:24:01 GMT)
Full text and
rfc822 format available.
Message #55 received at 75746 <at> debbugs.gnu.org (full text, mbox):
> From: Michael Albinus <michael.albinus <at> gmx.de>
> Cc: shuguang79 <at> qq.com, 75746-done <at> debbugs.gnu.org, shuguang.sun <at> zaiming.com
> Date: Fri, 14 Feb 2025 15:29:49 +0100
>
> Version: 30.2
>
> >> If I call the very same command in a native cmd.exe terminal, evrything
> >> works as expected. And so it did inside Emacs prior PuTTY 0.82.
> >
> > This was a deliberate behaviour change in 0.82, described in the
> > following link, which also describes how to restore the old behaviour.
> > https://www.chiark.greenend.org.uk/~sgtatham/putty/wishlist/windows-stdio-prompts.html
> >
> > The proposed option works as expected.
>
> I've adapted the call of plink, pscp, and psftp accordingly. This
> requires now at least PuTTY 0.82.
??? So users of older versions of PuTTY will now be unable to use
these Tramp methods, and _must_ upgrade their PuTTY installations? I
hope not.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 17:11:01 GMT)
Full text and
rfc822 format available.
Message #58 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Eli Zaretskii <eliz <at> gnu.org> writes:
Hi Eli,
>> > I'm not sure I understand the description of the new behavior: it says
>> > "(If there's no console involved, they'll fall back to the old
>> > behaviour.)", but in our case, i.e. invoking PuTTY from Emacs on
>> > Windows, there really isn't any console, since Emacs communicates with
>> > PuTTY via a pipe. Or am I missing something?
>>
>> Tramp opens a new local process, running cmdproxy.
>
> How does Tramp start cmdproxy? If that's via make-process or
> start-process, then the cmdproxy's standard input and output are
> connected to a pipe.
It's via start-process.
>> In its process buffer, it calls plink. So it might look like a
>> console from the PuTTY POV.
>
> I don't think so, see above.
>
> Perhaps PuTTY considers anything that is not a regular file a console?
> Otherwise, I don't understand why we see this issue.
Honestly, I have null knowledge about the topic on MS Windows.
> But if you are okay with forcing the new command-line option on plink,
> I don't mind. I'm just worried what will happen to older versions of
> plink when they are passed this option -- if they error out, we'll
> need to detect the version or ask users to configure it, which is not
> so good.
Yep, I'll work on it. See my other message to Robert.
Best regards, Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Fri, 14 Feb 2025 17:13:02 GMT)
Full text and
rfc822 format available.
Message #61 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Eli Zaretskii <eliz <at> gnu.org> writes:
>> I've adapted the call of plink, pscp, and psftp accordingly. This
>> requires now at least PuTTY 0.82.
>
> ??? So users of older versions of PuTTY will now be unable to use
> these Tramp methods, and _must_ upgrade their PuTTY installations? I
> hope not.
This is a temporary solution. There's enough time to improve it, until
end of this month (schedule of Tramp 2.7.2.2).
Best regards,Michael.
Information forwarded
to
bug-gnu-emacs <at> gnu.org
:
bug#75746
; Package
emacs
.
(Sat, 15 Feb 2025 16:28:02 GMT)
Full text and
rfc822 format available.
Message #64 received at 75746 <at> debbugs.gnu.org (full text, mbox):
Michael Albinus <michael.albinus <at> gmx.de> writes:
Hi Eli,
>>> I've adapted the call of plink, pscp, and psftp accordingly. This
>>> requires now at least PuTTY 0.82.
>>
>> ??? So users of older versions of PuTTY will now be unable to use
>> these Tramp methods, and _must_ upgrade their PuTTY installations? I
>> hope not.
>
> This is a temporary solution. There's enough time to improve it, until
> end of this month (schedule of Tramp 2.7.2.2).
FTR, I've adapted Tramp such a way that PuTTY 0.82 isn't a precondition
anymore. Pushed to master.
Best regards,Michael.
bug archived.
Request was from
Debbugs Internal Request <help-debbugs <at> gnu.org>
to
internal_control <at> debbugs.gnu.org
.
(Sun, 16 Mar 2025 11:24:17 GMT)
Full text and
rfc822 format available.
This bug report was last modified 93 days ago.
Previous Next
GNU bug tracking system
Copyright (C) 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson.