I'd like to move Tramp to SourceForge in the foreseeable future.
There are going to be changes to our network (firewall installation)
which will make it more difficult to allow others access.
Those of you who help develop Tramp might wish to register at
SourceForge so that you continue to have
[EMAIL PROTECTED] (Kai Großjohann) writes:
I'm now mass-subscribing you all to the Tramp mailing list on
SourceForge.
I'm less than impressed. A large text field titled ``Mass Subscribe
Members'' with an explanation of ``Enter one address per line''
somehow suggested that I could enter more
Each of you should have received the First SourceForge message a few
seconds ago. If not, please holler: I probably forgot to subscribe
you. Subscribing you was not automatic.
kai
--
Symbol's function definition is void: signature
Skip Montanaro [EMAIL PROTECTED] writes:
Now, about that filename format change... ;-)
What about it? Is the new format awful? I'd be willing to use
another format, but the format should make sure that it doesn't clash
with ange-ftp. Suggestions?
kai
--
Symbol's function definition is
[EMAIL PROTECTED] (Kai Großjohann) writes:
I'd like to move Tramp to SourceForge in the foreseeable future.
There are going to be changes to our network (firewall installation)
which will make it more difficult to allow others access.
I'm now mass-subscribing you all to the Tramp mailing
I have changed the default file name format.
Old:/r:user@host:/path/to/file /r@meth:user@host:/path/to/file
New:/./user@host:/path/to/file /./@meth:user@host:/path/to/file
This one should avoid clashes with ange-ftp, and also with the drive
letters on Windows.
Please test.
On Tue, 31 Jul 2001, Skip Montanaro wrote:
Compilation of tramp.el failed with this output:
xemacs -batch --eval (add-to-list 'load-path \.\) \
-f batch-byte-compile tramp.el
Compiling /home/skip/emacs/tramp/lisp/tramp.el...
While compiling toplevel forms in file
On 30 Jul 2001, Joe Wells wrote:
I visit a file in an unwritable remote directory. I modify the
buffer. I try to save my changes. The operation fails. The only
error message is tramp_exit_status 2 in the *tramp/METHOD HOST*
buffer.
Turning on tramp-debug-buffer reveals that what
On Tue, 24 Jul 2001, Ed Avis wrote:
The web page says 'The performance of the external transfer methods
is generally better than that of the inline methods.'.
Hm. Why does the web page say that? Hm. I think I didn't dare take
it out since I didn't really know it was false. And I've never
On Tue, 24 Jul 2001, Ed Avis wrote:
Could TRAMP include support for fsh
http://www.lysator.liu.se/fsh/? This is an rsh-compatible command
that reuses an ssh tunnel. In other words, once the ssh connection
has been established, it is left open for subsequent commands rather
than setting up
On Sun, 15 Jul 2001, Kai Großjohann wrote:
Once upon a time, I forked a `stable' branch. Then I made changes
to the stable branch as well as the head. Now I want to say:
whatever is in the head that seems stable enough, so just update my
stable branch to whatever is in the head.
Okay. I
I think I've managed to shoot myself in the foot with CVS. I wonder
if there is a doctor here who can help me with this kind of injury?
Once upon a time, I forked a `stable' branch. Then I made changes to
the stable branch as well as the head. Now I want to say: whatever is
in the head that
The host name of the CVS server has changed. It used to be bonny, now
it's cirian. This means that a `cvs update' will fail.
Maybe it's easiest to check out again into a fresh directory.
Else you might have success with editing all the CVS/Root files to
change the host name.
kai
--
Brian is reporting that Emacs and/or Tramp hangs when saving (not
loading) large files via an inline method. I'm wondering whether
maybe process-send-region has problems for large files, depending on
the OS?
Brian connects via the sm method to an HP-UX machine, I'm not sure but
guess that the
On Thu, 28 Jun 2001, Brian M. Fahs wrote:
It appears that I can not save a file to a remote HP-UX machine.
Here is the debugging output. I had to add the 1 sec. timeout
because it was hanging.
Maybe 1 second is too short. Can you add a timeout of, say, 60
seconds?
kai
--
~/.signature: No
On Wed, 27 Jun 2001, Spencer L. Swift wrote:
KAI -- If you read this soon, I still have the file open and have
done nothing to it. Is there anything I can do in a live session to
help debug this?
What does M-: (file-readable-p FILE) RET say?
What does M-: (file-writable-p FILE) RET say?
Now that you have a workaround, we might want to leasurely get to the
bottom of this. So I suggest that you get the most recent version via
CVS (but if you're afraid of bleeding-edge code, you could take the
last released version -- it's almost the same anyway :-). And then
you say (setq
I get confused. Hm. When running through Tramp, the remote end
doesn't send an answer for `cd ~; pwd', but it works when run outside
of Emacs.
Is the failure always the same, always at that spot?
kai
--
~/.signature: No such file or directory
On Mon, 18 Jun 2001, Brian M. Fahs wrote:
Waiting in wait-for-output.
It seems that you are using tramp2? (The directory contains both
tramp.el and tramp2.el.) Hm. Dunno what to do about tramp2.
Daniel? Should we use Brian as a guinea pig for testing Tramp2? Or
should we point him to the
On Mon, 18 Jun 2001, BRIAN FAHS wrote:
It seems to hang after it sends the mime-decode command. It never
returns from tramp-wait-for-output
Can you (setq tramp-debug-buffer t), then reproduce the problem, then
mail me (or the list) the *tramp/foo* and *debug tramp/foo* buffers?
Thanks,
kai
On 15 Jun 2001, Kai Großjohann wrote:
Okay, I have now put that version on the CVS server, too.
^^^
FTP
I mean FTP server, not CVS server.
kai
--
~/.signature: No such file or directory
On Fri, 15 Jun 2001, Tatsuo Natsukawa wrote:
Kai, you are right: the current CVS is ok. Thank you! Tatsuo
Okay, I have now put that version on the CVS server, too. Thanks for
testing.
kai
--
~/.signature: No such file or directory
On Thu, 14 Jun 2001, Tatsuo Natsukawa wrote:
Couldn't find exit status of `nil'
This means that Tramp was looking for some program on the remote host
(I think it was the `ls' program), but due to a timing problem, it
couldn't find the program name.
Can you do (setq tramp-debug-buffer t),
Ah, that bug. I think it is fixed in current CVS, could you please
test that? If your test says the current CVS is okay, then I'll
update the stable version.
The CVS instructions are part of the info file as well as the Web
page.
kai
PS: It seems that maintaining devel and stable releases is
Just in case someone feels inclined to implement this. Here is an
idea how to deal with the remote shell hanging up on us because of an
idle timeout. The problem right now is that a command might hang
because Emacs doesn't know that the remote end of an ssh connection
(say) has closed the
On 05 Jun 2001, Mark A. Hershberger wrote:
Gnus automatically senses when my IMAP connection times out and
reconnects. Does anyone here know enough about how it works to
say if we could duplicate its functionality?
I'm guessing (but only guessing) that a direct network stream goes to
state
On 04 Jun 2001, Stefan Monnier wrote:
This is because VC does not understand that your local userid is
really the same user as your remote userid and it then decides to
make the buffer read-only even though you have write access to it
(thinking you shouldn't modify a file that belongs to
On 03 Jun 2001, Daniel Pittman wrote:
Generally, using `verify-visited-file-modtime', as I
understand. Since tramp does not implement this, I don't know what
Emacs would do...
Hm. H Hmm...
I'm really confused. For it works fine if I enter a letter, it only
On Thu, 03 May 2001, [EMAIL PROTECTED] wrote:
At times, I will save a file (point a), make a few changes, and
then save the file again (point b). I'll then realize that the
changes were a complete mistake and C-_ back to point a without
saving (which I shall call point c). For some
I have recently made a change in the development version which might
help. I just declared the development version stable. Please try the
new tarball, available from the usual place:
http://ls6-www.cs.uni-dortmund.de/~grossjoh/emacs/tramp.html
Thanks,
kai
--
~/.signature: No such file or
I don't know if this helps, but I've made Tramp send Ctrl-D now. New
version in CVS. Does this change anything?
kai
--
~/.signature: No such file or directory
On Wed, 16 May 2001, Sridhar Boovaraghavan wrote:
However, I get a really strange problem. It tries to set the current
directory of the opened file to r: which doesn't exist. This then
causes problems with other buffers also complaining about not being
able to set the directory to r:
Argh.
On Wed, 30 May 2001, David Forrest wrote:
While using C-x f /r@sm:mug:docs/Quotes which is a 200 byte text
file I get an error of Lisp nesting exceeds max-lisp-eval-depth
Can you remove the Tramp *.elc files, then M-x toggle-debug-on-error
RET, then repeat the error? You should get a
On 27 May 2001, Stefan Monnier wrote:
Maybe you're victim of a strange stty setting ?
(there has been tty settings in the past where @ was set to `erase')
If that's the case, maybe Tramp should do a `stty erase ^?' ?
I have now done this. New version in CVS. Does this help, Sven?
kai
--
On 26 May 2001, Daniel Pittman wrote:
Maybe it's some very strange shell quoting bug in that version of
his shell or something?
Well, the Perl error message shows Perl code that looks like the
original source. Hm. But still, maybe the wrong quote is somewhere
else.
Sven?
kai
--
On Fri, 25 May 2001, Sven Utcke wrote:
$ tramp_file_attributes () {
/bin/perl -e '$f = $ARGV[0];
@s = lstat($f);
if (($s[2] 017) == 012) { $l = readlink($f); $l = \$l\; }
elsif (($s[2] 017) == 04) { $l = t; }
else { $l = nil };
printf((%s %u %u %u (%u %u) (%u %u) (%u %u)
On Fri, 25 May 2001, Sven Utcke wrote:
Right. Any difference in environment between the two?
kogs12/tmp% diff env.tramp env.ssh
LC_TIME, PATH and TERM could be further investigated. Can you make
the problem appear in a `normal' (non-Tramp) ksh by frobbing these
variables?
kai
--
On Thu, 24 May 2001, Matthew Harrison wrote:
Could you try this patch? If it works fine, I will commit it.
--- tramp.el.orig Thu May 24 17:30:22 2001
+++ tramp.elThu May 24 17:30:47 2001
@@ -3802,7 +3802,7 @@
(tramp-send-command multi-method method user host
Once upon a time, people had Perl 4 and Perl 5 on their system, and
invoking perl5 ran the newer version. I think these days this is
obsolete and Tramp should not check for perl5 anymore. What do you
think?
(Should Tramp assume that any Perl found is a Perl 5?)
kai
--
~/.signature: No such
On Fri, 25 May 2001, Sven Utcke wrote:
asterix/home/utcke% ls -l /bin/perl
lrwxr-xr-x1 root root 20 May 16 16:12 /bin/perl -
/usr/sbin/perl-5.004*
But the shell function calls /bin/perl5, not /bin/perl. Is that the
problem?
kai
--
~/.signature: No such file or directory
On Fri, 25 May 2001, Sven Utcke wrote:
(Should Tramp assume that any Perl found is a Perl 5?)
Hmm. This, of course, would be guaranteed to break once perl6 is
out, and possibly on some (very few) perl4-only machines...
I guess it's time to run `perl -v', then. Argh. Sounds like work.
On 25 May 2001, Stefan Monnier wrote:
Sven == Sven Utcke [EMAIL PROTECTED]
writes:
But the strange thing of course is that it _does_ work with that
particular instance of perl when called from the shell, just not
when called from tramp. This really _is_ weird...
Could it be something
On Wed, 23 May 2001, Chris Majewski wrote:
Does tramp have any support for remote debugging? So if I say M-x
gdb file in a remote buffer I get a remote gdb session? -chris
Not that I know of. Can you find out if that's supported at all, then
tell me how it is done with ange-ftp (if
On Wed, 23 May 2001, bruce ingalls wrote:
This is a good one to add to the FAQ:
M-x debug-on-entry
M-x efs-ftp-path
Maybe this is true for all advised functions?
kai
--
~/.signature: No such file or directory
On Thu, 24 May 2001, Chris Majewski wrote:
Looks like tramp is not sourcing my .cshrc, because I don't get the
PATH I want. How can I fix this? -chris
This is by design. Tramp needs a standard shell setup, and so if it
runs your shell init files -- bad. Even with a standard shell setup,
On Tue, 22 May 2001, Chris Majewski wrote:
1. Where is tramp-compile? My emacs finds no match for M-x
tramp-compile. This is $Id: tramp.el,v 2.0.2.3 2001/03/28 21:44:30
grossjoh Exp $
Do you have a file tramp-util.el? If not, then maybe I should do
something about the stable version...
On Wed, 23 May 2001, Chris Majewski wrote:
On 23 May 2001, Kai [iso-8859-1] Großjohann wrote:
On Wed, 23 May 2001, Chris Majewski wrote:
Cool, that worked. I had to put (require 'tramp-util) in my
.emacs. Now, how can I tell C-x C-e to run tramp-compile in a
remote buffer, and
On Wed, 23 May 2001, Chris Majewski wrote:
Symbol's value as variable is void: compilation-read-command
unless I do a M-x compile first (or run the function in a local
buffer).
Put (require 'compile) in your ~/.emacs.
kai
--
~/.signature: No such file or directory
On Mon, 21 May 2001, bruce ingalls wrote:
I regret, if someone replied; I may have lost some email during a
recent backup.
I'm sorry that I have been unable to keep up with my mail in the past
several months. :-(
EFS.el is ignoring Tramp's Advice() in the recent XEmacs 21.4.0.
How can I
On Wed, 23 May 2001, Chris Majewski wrote:
Cool, that worked. I had to put (require 'tramp-util) in my .emacs.
Now, how can I tell C-x C-e to run tramp-compile in a remote buffer,
and compile otherwise?
(defun cm-compile-maybe-remote ()
(interactive)
(if (and (buffer-file-name)
On Wed, 23 May 2001, Chris Majewski wrote:
On the other hand, every once in a
while I get an error like the following (the last line is the
meaningful one).
Argh. Looks like a timing problem. Please do (setq
tramp-debug-buffer t), wait for the problem to reappear, then have a
look at the
On Tue, 1 May 2001, Michael C. Adler wrote:
The result is the output from each test winds up being associated
with the wrong test and, ultimately, future shell commands:
In the development version, there is this change:
/
| revision 2.15
| date: 2001/04/13 16:35:07; author: grossjoh;
On 01 May 2001, Norbert Koch wrote:
C-x C-f /r@scp:db.LF.net:~/ork/perl/bla.pl
I have now changed Tramp to allow uppercase letters in user and host
names. New version in CVS. Does it work?
(I didn't test it -- I'm on vacation and so I'm allowed to be lazy :-)
kai
--
The passive voice
On 30 Apr 2001, Kirk Strauser wrote:
sh-2.05$ Enter passphrase for RSA key 'localuser@workstation':
Last login: Mon Apr 30 15:25:43 2001 from workstation on pts/8
Linux remoteserver1 2.2.17 #21 SMP Wed Aug 23 11:10:47 CDT 2000 i686 unknown
On 23 Apr 2001, Mark A. Hershberger wrote:
Looking through the debug buffer, it seems to be coming from the
following two lines:
ls: tramp_exit_status 1
/ this file does not exist : No such file or directory
The results of the two commands in
ls / this file does not
On Mon, 23 Apr 2001, Tom Roche wrote:
No, just more questions :-) As previously mentioned, but unanswered
:-(
I know. Things are becoming more and more difficult. It seems that
the easy bugs have all been fixed, now I'm confronted with toughies.
Especially on NT, I have no idea whatsoever
On 12 Apr 2001, Harry Putnam wrote:
That may be, I don't have a full samba install. Just enough to
access the smbclient. It acts very much like an ungarnished ftp
client, from what I've seen. No actual mounting like its on your
own machine. Just an ftp like interface, where you can
On Thu, 12 Apr 2001, Tom Roche wrote:
I can try this out. Do I need to do anything other than CVS
tramp.el?
Well, it might be nice if you get the rest of Tramp also from CVS.
Then you get the new manual and stuff. (Not that much has changed.)
Thanks for testing.
kai
--
Be indiscrete. Do
Could you (setq tramp-debug-buffer t), repeat the error, and mail the
contents of the *debug tramp/foo* buffer? Take care to remove all
passwords from the buffer. (I don't think any will be in it, but just
to be sure.)
What happens when you manually enter the value of $homepage, rather
than
On Wed, 11 Apr 2001, Stefan Monnier wrote:
Oh no! Here-documents!
Kai, is there a good reason for using here documents here ?
Maybe there was, once upon a time. I have now changed it and did a
quick test. Seems to work.
New version in CVS. Please test.
kai
--
Be indiscrete. Do it
On 12 Apr 2001, Stefan Monnier wrote:
On the other hand `HISTFILE=/dev/null /bin/ksh' will not work if the
user's SHELL is not bourne-like, which is more annoying. And I don't
think we can rely on /usr/bin/env being available, so it's not clear how
we can reliably set the env var before
On Thu, 12 Apr 2001, Tom Roche wrote:
[EMAIL PROTECTED] 12 Apr 2001 17:30:49 +0200
For example, what happens if we do "HISTSIZE=0" in addition to
"unset HISTFILE"?
IIRC I tried that, and it didn't work: ksh made temp files.
What was their size? So should HISTFILE be left at a default
On 12 Apr 2001, Harry Putnam wrote:
Has anyone every fiddled with getting tramp to work over smb
protocol?
Tramp wants a login shell and expects to be talking to a Unixish
system on the remote end. You could install an sshd and a bash on the
remote system, and if it looks sufficiently
On Tue, 10 Apr 2001, Tom Roche wrote:
? (Its determination would make a useful Doktorarbeit, however :-)
He, he. Not quite, I'm afraid. I wouldn't want to put so much effort
into it. Just choose some value, then let the users decide which is
best.
kai
--
Be indiscrete. Do it
On Mon, 09 Apr 2001, Tom Roche wrote:
It appears that tramp closes its ssh connection to a given host when
the last buffer connected with that host is killed.
I think Tramp never closes connections.
But Tramp could try to keep connections open, thwarting the idle
timeout imposed by ssh,
On Thu, 29 Mar 2001, Rafael Sepulveda wrote:
I want to use modified methods in tramp. Tha main reason is to use
them running '/bin/sh'so that the prompt that I use doesn't disturb
tramp (because of ansi escapes).
First of all, I have recently added to Tramp that it sets $TERM to
dumb while
I have now (finally) update the FAQ section to point to Tom's great
Web page. (Originally, I wanted to integrate the information into the
main body, but...)
This is in the development version you can get via CVS.
kai
--
Be indiscrete. Do it continuously.
Thanks, Tom, I've now changed the documentation a bit. It's in the
development version (available via CVS).
kai
--
Be indiscrete. Do it continuously.
On Tue, 27 Mar 2001, Kim Taylor wrote:
Basically, the admin command was passed
-i/r@METHOD:HOST:/PATH/TO/FILE/ where -iFILE would suffice.
Compounding the problem, the error wasn't noticed and the file and
buffer destroyed.
Could you M-x debug-on-entry RET vc-do-command RET and show me the
Argh. And why did I forget to tell you to do (setq tramp-verbose 10)?
Stupid me! I will make do with this, though. Sorry.
kai
--
Be indiscrete. Do it continuously.
Another idea: can you log in to the remote host, start ksh, then type
this command:
cd /non-existing-directory ; echo foo
Is `foo' printed? I'm suspecting that the shell ignores it if the
`cd' on the same line fails.
kai
--
Be indiscrete. Do it continuously.
On Wed, 28 Mar 2001, Mads Ipsen wrote:
So TRAMP tries to cd to a file, instead of opening it?
Well, err. Hm. Normally, it should `cd' to the directory containing
the file. It does `cd /foo; ls -l bar' rather than `ls -l /foo/bar'
because the output format is different. I think.
Can you
On 28 Mar 2001, Stefan Monnier wrote:
Why doesn't Tramp use
cd foobar echo foo
Because it wants to know the exit status of `cd', whether it failed or
not. (`foo' is actually `$?'.)
For some reason, however, I don't think that's it. Hm. If only I
knew what was going on. I think
Please try the new version from CVS or the ftp directory.
http://ls6-www.cs.uni-dortmund.de/~grossjoh/emacs/tramp.html
If that wasn't it, I'll add more debugging calls to tramp-message.
Maybe that helps to find it.
kai
--
Be indiscrete. Do it continuously.
On 27 Mar 2001, Stefan Monnier wrote:
"Benjamin" == Benjamin Rutt [EMAIL PROTECTED] writes:
I have not frobbed the backup-by-copying variable or any others.
The default value appears to be nil on my system, which would
suggest that renaming is used (and therefore modes should be
preserved).
Added to FAQ. Thanks.
kai
--
Be indiscrete. Do it continuously.
On Tue, 20 Mar 2001, Tom Roche wrote:
[EMAIL PROTECTED] 19 Mar 2001 20:52:13 +0100
What happens when you do it the second time?
Unfortunately, .sh_history doesn't stay dead:
Maybe there is a misunderstanding here. Doing the settings tells the
currently running instance of ksh to not keep
On Wed, 21 Mar 2001, Pete Forman wrote:
0 and 1 are non-starters. .sh_history is shared by all instances of
ksh. Deleting it will lose the history of any other active shells.
That's right. So Tramp just shouldn't write anything to the file.
It's bad enough at the moment that tramp swamps
I also think it's a timing problem. (setq tramp-debug-buffer t), post
the contents of *tramp/foo* in addition to *debug tramp/foo*.
kai
--
Be indiscrete. Do it continuously.
On Mon, 19 Mar 2001, Tom Roche wrote:
[EMAIL PROTECTED] 19 Mar 2001 20:52:13 +0100
What happens when you do it the second time?
I'm not sure what you mean--a second time in the same session? or
when I start a new session?
ksh
unset HISTSIZE
unset HISTFILE
foo
bar
# check .sh_history here
On 19 Mar 2001, Colin Marquardt wrote:
a small suggestion: since tramp is about the only thing which can
complete file names with remote connections, could it complete the
known host names from /etc/hosts too?
Good idea. We should set up a variable containing a list of hosts,
and then we
On Sun, 18 Mar 2001, Tom Roche wrote:
Truly annoying. What part of "no history" doesn't it understand ?-)
Or, now that I think of it ... could this actually be something
AFS is doing (given that 'afs' is part of the filename)? But why?
I don't know AFS. But you get similar files with NFS.
On Mon, 19 Mar 2001, Tom Roche wrote:
So, although 'unset' doesn't make .sh_history, or its successor, go
away upon _utterance_, it does appear to make it go away upon
_exit_-- which IIRC is better than what we had, and mostly solves
the problem of the ever-growing .sh_history (except for
On 19 Mar 2001, Christian Nyb wrote:
Is there a way to make tramp get my remote username from
.ssh/config?
Tramp is supposed to do this. Is your version old, perchance?
kai
--
Be indiscrete. Do it continuously.
On Mon, 19 Mar 2001, Tom Roche wrote:
Could this be added to tramp-util? It certainly adds utility for NT
trampers, and (AFAIK) testing
I think that Eli is right with his warnings, but I've now put adding
something like this on the todo list. You will have to call a
function to set this up,
On Sat, 17 Mar 2001, Tom Roche wrote:
[EMAIL PROTECTED] 18 Mar 2001 00:11:04 +0100
Umm ... I may not be such a skilled user, but I _do_ know how to
read a manpage :-)
*blush*
On Sat, 17 Mar 2001, Tom Roche wrote:
http://www.ntua.gr/cgi-bin/man-cgi?ksh+1
Command Re-entry
The text of the
On 18 Mar 2001, Alexnader Kotelnikov wrote:
Out of band method `scp' not applicable for remote shell asking for
a password
The solution is to use an inline method, such as `sm'. Is this
sufficient for you?
kai
--
Be indiscrete. Do it continuously.
I almost forgot to mention that, tomorrow, our machines will be down
_again_. Some disk was acting up which contains /usr for most of our
machines and /var/mail and other such stuff.
Our sysadmin is talking about a 2h downtime if everything works well
(SCSI bus too long, fixed by shorter cable)
On Sat, 17 Mar 2001, Tom Roche wrote:
On Fri, 16 Mar 2001, Tom Roche wrote:
I get
*vc-diff*
/bin/ksh: cvs: not found
How do I make ksh find cvs?
Also, just wondering: why ksh? (A quick hand-waving explanation will
probably suffice :-)
Well, Tramp works by sending commands to the
I think Tramp fails on all symlinks, but I haven't had time to
investigate this, yet. If this is the problem, then at least I know
that there is only one but, not two.
kai
--
Be indiscrete. Do it continuously.
On Sat, 17 Mar 2001, Tom Roche wrote:
* is tramp writing to this file?
No, Tramp itself is not writing to the file, but I think the shell
is. If you read the documentation for ksh and tell me how to turn off
the history, I'll put that code in Tramp. (There is already code for
turning off the
On Sat, 17 Mar 2001, Tom Roche wrote:
[EMAIL PROTECTED] 17 Mar 2001 13:32:21 +0100
Tramp does not know that this is your user id. There is a problem
with interoperation between VC and Tramp in this regard. Not sure
what to do about it.
Ach. Thanks anyway, [EMAIL PROTECTED]
Well, if
On Sat, 17 Mar 2001, Tom Roche wrote:
On Sat, 17 Mar 2001, Tom Roche wrote:
* is tramp writing to this file?
[EMAIL PROTECTED] 17 Mar 2001 21:18:10 +0100
No, Tramp itself is not writing to the file, but I think the shell
is. If you read the documentation for ksh and tell me how to turn
On 15 Mar 2001, Mark A. Hershberger wrote:
(if (not (nth 1 found))
! (setq found t)
(tramp-message 9 "Sending password...")
Eeek. This is bad. Could you try the new version from CVS (devel
branch, ie head, not stable branch) to see if that works?
I've adopted a
On Fri, 16 Mar 2001, Tom Roche wrote:
works for me, and seems even cleaner. (Except for the paren
placement: being usually procedural I like starts and ends to line
up :-)
I fail to see why people want to do this parens thing. Whoever looks
at them?
I just look at the indentation to see
On Fri, 16 Mar 2001, Tom Roche wrote:
and I know it works, but I don't understand why. However, _that_ is
not my immediate concern. Instead, I'd like to know: can TRAMP be
hacked to get a password itself, and pass it to the client via its
args (e.g. via tramp-telnet-args)? I.e. broken-client
On 16 Mar 2001, Daniel Katz wrote:
With a local file, emacs would notice that the file had changed on
disk and offer to re-read it into the buffer. Files that I visit by
tramp don't notice that the file has changed and don't ask whether I
want to refresh the buffer.
I wonder how does
On 15 Mar 2001, Daniel Pittman wrote:
True. I intend to take this up with the maintainers of it,
eventually. It's also true, though, of Ange-FTP.
To my knowledge, Ange-FTP isn't so pushy. For Ange-FTP, it's
sufficient to add it to file-name-handler-alist, and then to deal with
the functions
On 15 Mar 2001, Daniel Pittman wrote:
Anyway, I will look at that as a solution to the hook ordering
thing, unless someone felt like doing it for me. :)
Well, your current filename suggestion means there is no clash. Which
is good.
In case there is a filename syntax with clash, we can start
On Thu, 15 Mar 2001, [EMAIL PROTECTED] wrote:
debug:
Signaling: (wrong-type-argument listp t)
tramp-multi-connect-su(#process *tramp/multiu
ssh#mah@localhost:sudo#root@localhost:* "sudo" "root" "localhost" "sudo -u %u -s%n")
tramp-open-connection-multi("multiu" ["ssh" "sudo"] ["mah"
1 - 100 of 527 matches
Mail list logo
